Ethical Cognition, Anxiety, and Attitudes toward Artificial Intelligence in Higher Education: Validation and Predictive Modelling of the Albanian GAAIS

Ethical Cognition, Anxiety, and Attitudes toward Artificial Intelligence in Higher Education: Validation and Predictive Modelling of the Albanian GAAIS | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Ethical Cognition, Anxiety, and Attitudes toward Artificial Intelligence in Higher Education: Validation and Predictive Modelling of the Albanian GAAIS Elona Hasmujaj, Elvisa Drishti This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8483126/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 10 You are reading this latest preprint version Abstract Artificial intelligence is increasingly embedded in higher education, shaping teaching, assessment, and institutional governance. While existing research often focuses on perceived usefulness and performance, less attention has been paid to how ethical evaluations and emotional responses jointly shape attitudes toward artificial intelligence. This study examines attitudes toward artificial intelligence in Albanian higher education through an integrated cognitive–affective–ethical framework, focusing on ethical cognition and artificial intelligence–related anxiety as key determinants of positive and negative attitudes. A cross-sectional survey was conducted among 705 students and academic staff at Albanian universities. Three validated instruments were translated and culturally adapted: the General Attitudes toward Artificial Intelligence Scale, the Attitudes toward Ethical Artificial Intelligence Scale, and the Artificial Intelligence Anxiety Scale. Confirmatory factor analyses were performed to assess factorial validity and reliability. Hierarchical multiple regression models were then estimated to examine the predictive roles of ethical attitudes and artificial intelligence–related anxiety on positive and negative attitudes, controlling for sociodemographic characteristics. Confirmatory factor analyses supported the original factorial structures of all three instruments, with excellent model fit and strong internal consistency. Ethical cognition and emotional responses jointly explained substantial variance in attitudes toward artificial intelligence. Among ethical dimensions, non-maleficence was the only factor consistently associated with more positive attitudes, whereas privacy concerns significantly increased negative evaluations. Both cognitive and affective forms of artificial intelligence anxiety reduced positive attitudes and amplified negative ones, with affective anxiety showing the strongest effects. Ethical and emotional factors together explained 47% of the variance in positive attitudes and 39% in negative attitudes. Attitudes toward artificial intelligence in higher education are shaped by an interplay of ethical evaluation and emotional response rather than by instrumental considerations alone. Ethical reassurance related to harm avoidance and effective management of anxiety appear central to fostering constructive engagement with artificial intelligence. These findings highlight the importance of integrating ethical reflection and emotional preparedness into artificial intelligence literacy initiatives, particularly in transitional higher-education contexts. Artificial intelligence higher education Albania ethical cognition AI anxiety technology acceptance 1. Introduction The rapid diffusion of artificial intelligence (AI) across educational, professional, and social domains has intensified scholarly interest in how individuals perceive, evaluate, and emotionally respond to intelligent systems (Bankins and Formosa, 2023 ; Schepman and Rodway, 2025 ; Sultana et al., 2025 ). Within higher education, AI tools are increasingly embedded in teaching, assessment, and administrative processes, reshaping how knowledge is produced, delivered, and governed (Hubertz and Janowsky, 2025 ; Mubashir et al., 2025 ; Weng et al., 2024 ). While these developments promise gains in efficiency, personalisation, and accessibility, they simultaneously raise concerns related to privacy, fairness, accountability, and the preservation of human agency. To capture the multidimensional nature of these responses, the present study adopts an integrated cognitive–affective–ethical framework for analysing attitudes toward AI in higher education. This framework draws on three complementary theoretical traditions. First, the Attitude–Behaviour Model Model (Ajzen, 1982 ; Ajzen and Fishbein, 2000 ; Liska, 1984 ) conceptualises attitudes as composite evaluations arising from the interaction between cognitive appraisals (beliefs, judgments, evaluations) and affective reactions (emotions such as anxiety, curiosity, or enthusiasm). In this study, ethical cognition represents the cognitive dimension—individuals’ evaluations of fairness, transparency, non-maleficence, privacy, and responsibility in AI systems—while AI anxiety captures the affective dimension, encompassing both cognitive concerns (e.g., difficulty understanding AI) and emotional responses (e.g., fear of dependence or displacement) (Wang and Wang, 2022 ) Second, insights from the Affect Heuristic and dual-process theories of risk perception (Skagerlund et al., 2020 ; Slovic et al., 2007 , 2004 ) emphasise that emotions are not subordinate to reason but function as parallel processing channels that strongly influence judgments under conditions of complexity and uncertainty. In the context of AI, affective cues—such as fear, fascination, or unease—often shape perceptions of risk, trust, and legitimacy more powerfully than objective knowledge. Ethical reassurance (cognitive) and emotional anxiety (affective) therefore operate jointly in shaping overall evaluations of AI’s desirability and social acceptability. Third, the study extends insights from the Technology Acceptance Model (TAM) (Davis, 1989) and the Unified Theory of Acceptance and Use of Technology (UTAUT/UTAUT2) (Venkatesh et al., 2003 ), which traditionally link perceived usefulness and ease of use to technology adoption. Recent extensions of these models increasingly recognise that moral trust, perceived fairness, and emotional comfort play a central moderating role in technology acceptance, particularly for high-stakes and opaque technologies such as AI (Mubashir et al., 2025 ; Schepman and Rodway, 2025 ; Sultana et al., 2025 ). By integrating ethical cognition (measured through AT-EAI) and emotional response (measured through AIAS) as antecedents of general attitudes toward AI (measured through GAAIS), the present study situates AI acceptance within a broader socio-psychological and ethical framework rather than a narrowly utilitarian one. Growing empirical evidence supports this synthesis, showing that individuals’ attitudes toward AI are shaped as much by moral and emotional considerations as by instrumental evaluations of performance or (Lee et al., 2015 ; Longoni and Cian, 2022 ). Ethical judgments concerning fairness, transparency, and responsibility constitute a cognitive foundation for trust, while emotional responses such as anxiety or excitement represent the affective component of technological appraisal. Together, these dimensions determine whether AI is perceived as empowering or threatening. From an educational perspective, this inquiry is particularly salient in transitional contexts such as Albania, where digital transformation has advanced more rapidly than institutional regulation and ethical governance frameworks (Council of Ministers, 2025 ; GoA, 2025 ; MASR, 2021 ). Students and academics are increasingly exposed to AI-enabled learning management systems, plagiarism detection tools, and generative AI applications, often in the absence of systematic ethical guidance or robust data governance mechanisms (MAS, 2025 ). In such environments, perceptions of AI are shaped less by formal regulation and more by informal experiences, cultural narratives, and emotional reactions. Examining how ethical awareness and AI-related anxiety jointly influence attitudes toward AI in higher education therefore offers critical insights into digital readiness, ethical literacy, and the psychology of technology acceptance in emerging economies (EU4Youth, 2024 ; GoA, 2025 ). To address these issues empirically, the study employs three internationally recognised instruments. The Attitudes toward Ethical Artificial Intelligence Scale (AT-EAI) (Jang et al., 2022 ) assesses individuals’ endorsement of ethical principles guiding AI design and use. The AI Anxiety Scale (AIAS) (Wang and Wang, 2022 ) captures cognitive and affective anxiety elicited by AI technologies. The General Attitudes toward Artificial Intelligence Scale (GAAIS) (Schepman and Rodway, 2025 , 2020 ) measures both optimistic and sceptical orientations toward AI. Together, these instruments operationalise the cognitive, affective, and global components of attitude formation, providing an integrated framework for analysing how ethical cognition and emotional responses shape attitudes toward AI in higher education. Accordingly, this study pursues two objectives. First, it conducts a confirmatory validation of the Albanian versions of the GAAIS, AT-EAI, and AIAS, assessing their factorial structure and psychometric reliability. Second, it applies predictive modelling to examine how ethical attitudes and AI-related anxiety explain variation in positive and negative attitudes toward AI, controlling for demographic factors such as gender, age, education level, and exposure to AI-related information. This sequential design—combining confirmatory factor analysis with hierarchical regression—aligns with the methodological logic of the original GAAIS validation studies and allows for a robust assessment of both measurement and predictive validity in the Albanian higher-education context (Cicero et al., 2025 ; Hubertz and Janowsky, 2025 ). By linking ethics, emotion, and attitude formation, this study contributes to the expanding literature on the psychological and moral foundations of technology acceptance (Jang et al., 2022 ; Mubashir et al., 2025 ; Weng et al., 2024 ). It advances the field in three ways: first, by providing validated Albanian versions of widely used international instruments, enabling cross-cultural research; second, by empirically demonstrating the joint role of ethical cognition and emotional response in shaping positive and negative attitudes toward AI; and third, by offering policy-relevant insights for universities seeking to integrate AI responsibly through curricula that combine technical competence with ethical reflection and emotional preparedness. The remainder of the paper is structured as follows. Section 2 reviews the theoretical and empirical literature on ethical and emotional determinants of AI attitudes. Section 3 describes the data, instruments, and analytical strategy. Section 4 presents the empirical results. Section 5 discusses implications for theory, practice, and higher-education policy. Section 6 concludes with limitations and directions for future research. 2. Literature review 2.1. Measuring Attitudes Toward Artificial Intelligence: The GAAIS Framework The General Attitudes toward Artificial Intelligence Scale (GAAIS), developed by Schepman and Rodway ( 2020 ), represents one of the most widely adopted instruments for capturing the multidimensional nature of public attitudes toward AI. Rather than treating attitudes as a unidimensional continuum ranging from acceptance to rejection, the GAAIS explicitly distinguishes between positive and negative evaluative dimensions. This conceptualisation reflects growing recognition in the attitude and risk-perception literature that individuals often hold ambivalent or coexisting evaluations of emerging technologies, simultaneously acknowledging their benefits while expressing concern about their societal and personal consequences. The positive dimension of the GAAIS captures attitudes related to curiosity, enthusiasm, perceived usefulness, and anticipated societal or personal benefits of AI. Items loading on this factor reflect optimism regarding AI’s potential to enhance efficiency, improve well-being, and support human activities. In contrast, the negative dimension captures scepticism, distrust, and perceived risk, encompassing concerns about loss of control, ethical misuse, surveillance, and broader social harms. Importantly, these two dimensions are not conceptualised as simple opposites but as partially independent constructs, allowing individuals to express both high optimism and high concern simultaneously. The original validation study conducted in the United Kingdom and Italy reported a robust two-factor structure with strong psychometric properties, including high internal consistency for both the positive and negative subscales (Cicero et al., 2025 ; Sacco et al., 2025 ; Schepman and Rodway, 2020 ). Subsequent applications and validations across diverse sociocultural contexts – including East Asia, Latin America, and the Middle East - have largely replicated this bi-dimensional architecture, supporting the structural stability and cross-cultural relevance of the GAAIS (Lee et al., 2015 ; Longoni and Cian, 2022 ). These studies consistently show that general attitudes toward AI are best understood as a configuration of concurrent positive expectations and negative concerns, rather than as a single evaluative stance. At the same time, cross-national research highlights that the relative salience of positive and negative attitudes varies systematically across cultural, institutional, and technological contexts. In societies characterised by lower levels of institutional trust, limited regulatory oversight, or uneven exposure to advanced digital technologies, negative affective responses – such as anxiety, uncertainty, and perceived loss of agency – tend to be more pronounced (Hubertz and Janowsky, 2025 ; Kaya et al., 2024 ; Mantello et al., 2025 ). Conversely, in contexts where AI is embedded within stable governance frameworks and transparent institutional environments, positive evaluations related to utility and innovation are more likely to dominate. These insights underscore the importance of contextual validation when applying the GAAIS beyond the settings in which it was originally developed. This is particularly relevant for Albania, where attitudes toward automation, digital governance, and algorithmic decision-making are shaped by a combination of rapid digitalisation, ongoing institutional transition, and historically low levels of public trust in formal institutions. In such contexts, perceptions of AI may be more strongly influenced by affective responses, ethical concerns, and informal experiences than by established regulatory assurances. Validating the GAAIS in the Albanian higher-education context is therefore essential to ensure both its psychometric robustness and its conceptual sensitivity to local socio-institutional conditions. 2.2. Ethical Attitudes Toward Artificial Intelligence: The AT-EAI Scale The Attitudes toward Ethical Artificial Intelligence Scale (AT-EAI) developed by Jang et al. ( 2022 ) operationalises the normative and ethical dimensions of AI acceptance, shifting the focus from instrumental evaluations toward questions of moral legitimacy and societal responsibility. Grounded in international frameworks for trustworthy AI - most notably those articulated by UNESCO and the OECD - the AT-EAI conceptualises ethical attitudes as a set of cognitive evaluations concerning whether AI systems conform to widely endorsed moral principles. Specifically, the scale captures five interrelated ethical domains. Fairness refers to the expectation that AI systems should produce non-discriminatory outcomes and avoid bias against individuals or social groups. Transparency concerns the extent to which AI systems are explainable, interpretable, and open to scrutiny by users and affected stakeholders. Non-maleficence reflects the foundational ethical principle of “do no harm,” encompassing concerns about unintended negative consequences, misuse, or adverse societal impacts. Privacy addresses the protection of personal data and the responsible handling of information in AI-driven processes. Finally, responsibility relates to accountability mechanisms, including clarity over who is answerable for AI-generated decisions and outcomes. Empirical validations of the AT-EAI across diverse cultural settings - including Japan, South Korea, and several European countries - have demonstrated strong psychometric properties, with internal consistency coefficients typically ranging between α = 0.74 and α = 0.84, as well as stable factorial structures across contexts (Kong et al., 2023 ; Lim et al., 2025 ; Sun and Zhou, 2025 ; Yilmaz et al., 2024 ; Zhu et al., 2025 ). Beyond measurement reliability, a growing body of research indicates that perceived ethical alignment plays a central role in shaping trust, legitimacy, and acceptance of AI applications (Arora and Garg, 2023 ; Jobin et al., 2019 ). Ethical evaluations thus function as a critical cognitive filter through which individuals assess whether AI technologies are socially acceptable and worthy of integration into sensitive domains. In the context of higher education, ethical perceptions are particularly salient. Students and academic staff increasingly interact with algorithmic systems - such as plagiarism detection software, automated grading tools, learning analytics, and adaptive learning platforms - whose decision logic is often opaque and poorly understood (Jang et al., 2022 ; Sultana et al., 2025 ). These systems directly affect academic evaluation, surveillance, and data use, amplifying concerns related to fairness, transparency, and accountability. Validating the AT-EAI in the Albanian context therefore extends its applicability to a setting where formal awareness of AI ethics remains limited and regulatory frameworks are still evolving, making ethical cognition a potentially decisive factor in shaping attitudes toward AI in universities. 2.3. Anxiety Toward Artificial Intelligence: The AIAS Framework While ethical attitudes capture the cognitive–normative evaluation of AI, emotional responses represent an equally important component of attitude formation. The AI Anxiety Scale (AIAS) introduced by (Wang et al., 2025 ; Wang and Wang, 2022 ) addresses this affective dimension by measuring the extent to which AI technologies evoke feelings of apprehension, unease, or threat. The scale is theoretically grounded in dual-process models of emotion and risk perception, which distinguish between cognitive appraisals of threat and affective emotional reactions (Slovic et al., 2007 , 2004 ). The AIAS differentiates between two related but distinct forms of anxiety. Cognitive anxiety reflects concerns about understanding AI systems, keeping pace with rapid technological change, and coping with perceived skill obsolescence. It captures feelings of confusion, uncertainty, and perceived inadequacy in the face of increasingly complex AI technologies. Affective anxiety, by contrast, refers to emotional unease and fear, including worries about dependency on AI, loss of human autonomy, or displacement of human roles by intelligent systems (Skagerlund et al., 2020 ; Wang and Wang, 2022 ).. This two-factor structure has been consistently supported across empirical studies and aligns with established affective–cognitive models of technological anxiety. Psychometric evaluations of the AIAS report high internal consistency (typically α > .85) and stable factorial validity across Western and Asian samples (Kaya et al., 2024 ; Nomura and Tanaka, 2022 ; Wang et al., 2025 ; Wang and Wang, 2022 ). Substantive research using the scale shows that higher levels of AI-related anxiety are associated with lower perceived usefulness, reduced acceptance of AI technologies, and stronger ethical and social concerns. Conversely, exposure to AI education, transparent system design, and opportunities for informed engagement have been shown to attenuate anxiety and foster more favourable attitudes toward AI. In higher-education settings, AI anxiety is particularly consequential. Students are often required to engage with AI-enabled systems in evaluative and high-stakes contexts—such as assessment, academic integrity monitoring, and performance analytics—where uncertainty and perceived loss of control may heighten emotional responses. Recent evidence suggests that emotional reactions to AI can shape general attitudes as strongly as cognitive evaluations of ethics or utility (Kong et al., 2023 ; Xue et al., 2024 ). The AIAS is therefore well suited to examining how affective anxiety interacts with ethical cognition to shape positive and negative attitudes toward AI in higher education, particularly in transitional contexts such as Albania, where rapid digitalisation coexists with limited institutional reassurance. 2.4. The Albanian Context: Digitalisation, Youth, and Higher Education Digitalisation has emerged as a central pillar of Albania’s development agenda, framed simultaneously as an educational reform imperative and a strategic response to youth labour-market challenges. In the education sector, the National Strategy for Education 2021–2026 identifies digital competence as a core transversal objective (Objective A4), emphasising the systematic integration of information and communication technologies (ICT) across teaching, learning, and institutional management (GoA, 2025 ; MASR, 2021 ). This policy orientation reflects broader efforts to align Albania’s education system with European standards and to equip students with skills relevant to a rapidly transforming digital economy. Despite these ambitions, structural constraints continue to shape the lived realities of digitalisation in education. Persistent infrastructural gaps remain evident, particularly in earlier stages of the education system. As reported by the Council of Ministers ( 2025 ) and Ministry of Education and Sports (MAS, 2025 ), the computer-to-student ratio in pre-university education stood at approximately 1:27 in 2017, a figure markedly below the European Union benchmark range of 1:3 to 1:7. While conditions improve at the tertiary level, these disparities signal cumulative inequalities in digital exposure that accompany students into higher education. To mitigate such gaps, the Albanian Academic Network (RASH) plays a key enabling role by providing universities with access to learning management systems, digital repositories, research databases, and blended-learning infrastructures, thereby supporting the gradual digital modernisation of higher-education institutions (Council of Ministers, 2025 ). Recent national policy documents explicitly frame AI adoption around trustworthy and human-centred use, highlighting ethical safeguards, transparency, data governance, and alignment with emerging EU regulatory approaches. This macro-policy emphasis makes universities a critical implementation arena, where student acceptance is likely to hinge on whether AI systems are experienced as safe (non-maleficence) and rights-protective (privacy and accountability), rather than merely efficient. Against this background, examining ethical endorsement and anxiety as joint predictors of AI attitudes offers an empirically grounded indicator of Albania’s ‘readiness’ for responsible AI integration in higher education (GoA, 2025 ). Parallel developments are evident in the labour-market and youth-policy domains. The Youth Guarantee (YG) initiative, implemented in Albania with EU support, explicitly integrates digital tools to engage young people not in employment, education, or training (NEETs). The YG’s online platform - embedded within the national e-Albania portal - and the forthcoming dedicated Digital Portal aim to streamline registration, profiling, counselling, and referral to training or employment opportunities (EU4Youth, 2024 ; European Commission, 2025a , 2025b ). These efforts are reinforced by the National Youth Strategy 2022–2029, which prioritises digital literacy, online career guidance, and technology-mediated access to services as mechanisms for improving youth employability and social inclusion. In parallel, the expansion of digital freelancing and remote work has enabled segments of Albanian youth to participate in global labour markets, often bypassing domestic institutional constraints (ETF, 2022 ). However, the rapid diffusion of digital tools has not been matched by equivalent progress in institutional capacity, pedagogical preparedness, or ethical governance, particularly within higher education. Universities continue to face challenges related to uneven digital infrastructure, limited staff training in advanced digital and AI-enabled tools, and the absence of systematic frameworks for ethical reflection on algorithmic systems. As a result, students’ encounters with AI technologies - ranging from learning platforms and plagiarism detection software to generative AI applications - are frequently mediated through informal learning, peer experimentation, and fragmented institutional guidance. In this context, attitudes toward AI among Albanian students are likely to be shaped by a combination of unequal digital access, experiential uncertainty, and limited exposure to structured ethical discourse. Ethical awareness and AI-related anxiety thus become especially salient lenses through which to examine how digital transformation is cognitively and emotionally internalised by young people preparing to enter both national and transnational labour markets. Analysing these dimensions within the Albanian higher-education context offers valuable insights into the socio-psychological foundations of AI acceptance in emerging economies, where technological acceleration often outpaces institutional regulation and ethical governance. 2.5. Theoretical and Empirical Foundations for Predictive Modelling To account for the multidimensional structure of attitudes toward artificial intelligence, the present study adopts an integrated cognitive–affective–ethical framework that synthesises three complementary theoretical traditions: (i) affective–cognitive models of attitude formation, (ii) the affect heuristic and dual-process theories of risk perception, and (iii) extended technology acceptance models (TAM and UTAUT2). Together, these perspectives provide a coherent theoretical rationale for examining how ethical cognition and emotional responses jointly shape general attitudes toward AI and justify their inclusion as key predictors in the empirical model. First, affective–cognitive models of attitude formation conceptualise attitudes as composite evaluations emerging from the interaction between cognitive beliefs and affective reactions (Ajzen and Fishbein, 2000 ; Eagly and Chaiken, 1993 ; Liska, 1984 ). Attitudes are not reducible to either rational judgment or emotional response alone; rather, they reflect the dynamic integration of both components. In the present study, ethical cognition, operationalised through the Attitudes toward Ethical Artificial Intelligence Scale (AT-EAI), represents the cognitive evaluative dimension. It captures individuals’ reasoned assessments of AI systems in terms of fairness, transparency, non-maleficence, privacy, and responsibility. AI anxiety, measured by the AI Anxiety Scale (AIAS), represents the affective dimension of attitude formation, encompassing both cognitive anxiety (concerns about understanding and keeping pace with AI developments) and affective anxiety (emotional unease related to dependence, loss of control, or human replacement) (Wang et al., 2025 ; Wang and Wang, 2022 ). The interaction between these cognitive and affective components constitutes the attitudinal core reflected in the General Attitudes toward Artificial Intelligence Scale (GAAIS). Second, insights from the affect heuristic and dual-process theories of risk perception further clarify why ethical evaluations and emotional responses should be modelled jointly rather than independently. Dual-process frameworks posit that cognition and affect operate as parallel systems that jointly inform judgment, particularly under conditions of uncertainty and complexity (Slovic et al., 2007 , 2004 ). In such contexts, affective reactions often serve as heuristic shortcuts that shape perceptions of risk, trust, and acceptability more rapidly and sometimes more strongly than deliberative reasoning. AI technologies, characterised by opacity, autonomy, and moral ambiguity, are especially prone to affect-driven evaluation. Empirical research shows that emotions such as fear, unease, or moral concern can amplify perceived risk and suppress acceptance even when perceived benefits are acknowledged (Schepman and Rodway, 2020 ; Sultana et al., 2025 ). Consequently, ethical reassurance (as a cognitive signal of legitimacy) and AI-related anxiety (as an affective signal of threat or uncertainty) are expected to interact in shaping overall judgments of AI’s desirability and trustworthiness. Third, this study builds on and extends insights from the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT/UTAUT2), which traditionally explain technology adoption through perceived usefulness, ease of use, and facilitating conditions (Sun and Zhou, 2025 ; Venkatesh et al., 2003 ). While these frameworks have proven robust for explaining adoption of conventional information systems, recent empirical work highlights their limitations when applied to AI technologies, which raise broader ethical, social, and emotional concerns. Extensions of TAM and UTAUT increasingly incorporate trust, moral evaluation, and emotional comfort as central determinants of acceptance, particularly in high-stakes or socially sensitive domains (Longoni and Cian, 2022 ; Nomura and Tanaka, 2022 ). From this perspective, general attitudes toward AI are better understood as the outcome of an evaluative process that integrates ethical judgment and emotional response, rather than as a direct function of instrumental utility alone. Empirically, this integrated framework leads to clear expectations for predictive modelling. Individuals who perceive AI systems as ethically legitimate, fair, and accountable are expected to report stronger positive attitudes and weaker negative attitudes toward AI. Conversely, higher levels of AI-related anxiety, whether cognitive or affective, are expected to dampen optimism and amplify scepticism, risk sensitivity, and negative evaluations. Ethical cognition and emotional response are therefore conceptualised as dual antecedents of general attitudes toward AI, exerting independent and potentially opposing effects on positive and negative attitude dimensions. The resulting conceptual pathway underpinning the empirical analysis can be summarised as follows: $$\:Ethical\:Cognition\:\left(Cognitive\:Appraisal\right)\:+\:AI\:Anxiety\:\left(Affective\:Reaction\right)\:\to\:\:General\:Attitudes\:toward\:AI\:\left(Positive\:and\:Negative\:Evaluations\right)$$ Eq. 1 This theoretical synthesis provides the foundation for the study’s predictive modelling strategy, which empirically tests these relationships using hierarchical regression analyses following the confirmatory validation of the GAAIS, AT-EAI, and AIAS instruments. By embedding predictive modelling within a clearly articulated socio-psychological framework, the study moves beyond descriptive validation and offers a theory-driven explanation of how ethical and emotional mechanisms jointly shape attitudes toward AI in higher education. 3. Research Hypotheses Building on affective-cognitive models of attitude formation, dual-process theories of risk perception, and extended technology acceptance frameworks, the study formulates the following hypotheses concerning the ethical and emotional determinants of students’ attitudes toward artificial intelligence in higher education. H1a (Ethical cognition and positive attitudes). Stronger endorsement of ethical principles guiding AI (AT-EAI dimensions) is positively associated with Positive GAAIS scores, reflecting greater optimism and perceived benefits of AI. H1b (Ethical cognition and negative attitudes). Stronger endorsement of ethical principles guiding AI is negatively associated with Negative GAAIS scores, indicating reduced scepticism and perceived risk; however, privacy concerns may be positively associated with Negative GAAIS by heightening sensitivity to data-related risks. H2a (Cognitive anxiety and positive attitudes). Higher levels of cognitive AI anxiety are negatively associated with Positive GAAIS scores. H2b (Affective anxiety and positive attitudes). Higher levels of affective AI anxiety are negatively associated with Positive GAAIS scores. H3a (Cognitive anxiety and negative attitudes). Higher levels of cognitive AI anxiety are positively associated with Negative GAAIS scores. H3b (Affective anxiety and negative attitudes). Higher levels of affective AI anxiety are positively associated with Negative GAAIS scores. Together, these hypotheses reflect an integrated cognitive–affective–ethical model in which ethical evaluations and emotional responses function as distinct but complementary antecedents of both optimistic and risk-oriented attitudes toward artificial intelligence. 4. Methods 4.1. Participants and procedure This study employed a cross-sectional survey design conducted between February and April 2025 at the University of Shkodra “Luigj Gurakuqi,” Albania. Data were collected using a structured questionnaire administered both in-class and via an online Qualtrics link to ensure broad participation across study programmes. After screening for incomplete questionnaires and patterned or inattentive responses, a total of 705 valid observations were retained for analysis. Participants were drawn from a wide range of academic programmes, including business, education, information sciences, and social sciences, broadly reflecting the university’s enrolment structure. The sample was 71.4% female and 28.6% male, consistent with the gender composition of Albanian higher education, particularly in the social sciences and education fields (MASR, 2024). With respect to age, 56.9% of respondents were aged 18–25, 25.5% were 26–35, and 17.6% were 36–50, indicating the inclusion of both traditional students and mature learners. Approximately half of the respondents were enrolled at the undergraduate level, while master’s and other postgraduate students constituted the remainder. The Albanian higher-education context is particularly appropriate for examining attitudes toward artificial intelligence. Students and academic staff increasingly encounter AI-enabled tools-such as generative AI applications, plagiarism detection systems, and learning-management-system analytics-within an institutional environment where frameworks for data protection, algorithmic transparency, and AI ethics are still evolving. Participation was voluntary, informed consent was obtained prior to survey completion, and no incentives were offered. 4.2. Measures All constructs were measured using internationally validated instruments, translated and adapted into Albanian following standard translation–back-translation procedures to ensure conceptual equivalence. Unless otherwise specified, items were rated on five-point Likert scales (1 = strongly disagree, 5 = strongly agree). Ethical evaluations of AI were measured using the 17-item AT-EAI developed by Jang et al. ( 2022 ). The scale captures five core dimensions of trustworthy AI: fairness, transparency, non-maleficence, privacy, and responsibility. Example items include “AI should be designed to avoid bias or discrimination” (fairness) and “People should be informed when AI influences outcomes” (transparency). Reverse-coded items were recoded so that higher scores reflected stronger endorsement of ethical AI principles. In the present sample, internal consistency was satisfactory across all subscales (α = 0.74–0.83), consistent with prior validation studies. Emotional responses to AI were assessed using the AIAS (Wang et al., 2025 ; Wang and Wang, 2022 ), which consists of 14 items loading on two factors: cognitive anxiety (difficulty understanding or keeping up with AI developments) and affective anxiety (fear of dependence, loss of control, or role replacement). To preserve comparability with the original instrument and retain response variability, the original seven-point Likert scale (1 = strongly disagree, 7 = strongly agree) was maintained. Example items include “Learning to use AI products makes me anxious” (cognitive anxiety) and “I fear AI could replace human workers” (affective anxiety). Reliability was excellent (overall α = 0.91; cognitive α = 0.88; affective α = 0.86). General attitudes toward AI were measured using GAAIS scale, developed by Schepman and Rodway ( 2020 ). The Albanian version retained the original two-factor structure: (i) Positive GAAIS (12 items), capturing optimism, interest, and perceived societal benefits (e.g., “Much of society will benefit from a future with AI”); and (ii) Negative GAAIS (8 items), capturing discomfort, concern, and perceived danger (e.g., “AI is dangerous”). Both subscales demonstrated high internal consistency in the present sample (α = 0.89 for Positive; α = 0.83 for Negative), in line with confirmatory factor analysis results reported later. To isolate the predictive effects of ethical attitudes and AI anxiety, several background variables were included as controls: gender (1 = male), age group (18–25; 26–35; 36–50 as reference), educational status (bachelor student, bachelor graduate, master student, master graduate; PhD/other as reference), and self-reported knowledge of AI (no information, little, sufficient, extensive). Birth order (only child, first-born, middle, youngest) was also recorded to explore potential family-structure effects, but was treated as a secondary control. 4.3. Ethical Considerations The study was conducted in accordance with the ethical standards of the University of Shkodra “Luigj Gurakuqi.” Ethical approval was granted by the University Ethics Committee (Approval No. 2025/AEI-05). Participants were informed about the study’s objectives, assured of anonymity and confidentiality, and informed of their right to withdraw at any time without consequence. No personally identifying information was collected. All instruments underwent translation and back-translation procedures, and a pilot study with 35 students was conducted to assess clarity, cultural appropriateness, and completion time. Minor linguistic refinements—primarily simplifying items related to transparency and responsibility—were implemented prior to full deployment. Data were stored on password-protected institutional servers and used exclusively for research purposes. 4.4. Descriptive Analysis of the Sample Table 1 presents descriptive statistics for the 705 respondents included in the analysis. The demographic composition closely mirrors both the structure of the University of Shkodra “Luigj Gurakuqi” and broader participation patterns in Albanian higher education. A pronounced gender imbalance is evident, with women constituting 71.4% of the sample. This pattern aligns with national trends indicating female predominance in education, business, and social-science programmes (GoA, 2025 ; MAS, 2025 ; MASR, 2021 ), reflecting the broader feminisation of tertiary education in Albania. In terms of age, the majority of respondents (56.9%) were aged 18–25, representing traditional undergraduate students. A further 25.5% were aged 26–35, typically postgraduate students, while 17.6% were aged 36–50, capturing mature learners and doctoral candidates. This distribution enables generational comparisons in exposure to digital technologies and familiarity with AI. Educational status was heterogeneous: 30% were bachelor’s students, 14.7% bachelor graduates, 18% master’s students, and 34% master’s graduates, with 3.1% reporting doctoral-level engagement. This mix ensures representation across academic stages, from early users of digital learning systems to advanced researchers. Table 1 Descriptive Statistics of Sample Characteristics (N = 705) Variable Obs. M SD Min Max Male 705 0.286 0.452 0 1 Age group 18–25 years 705 0.568 0.495 0 1 26–35 years 705 0.255 0.436 0 1 36–50 years (ref. category) 705 0.175 0.381 0 1 Educational status Bachelor student 705 0.300 0.458 0 1 Bachelor graduate 705 0.147 0.354 0 1 Master student 705 0.180 0.384 0 1 Master graduate 705 0.340 0.474 0 1 PhD holder (ref. category) 705 0.031 0.174 0 1 Self-reported knowledge of AI No information (ref. category) 705 0.090 0.287 0 1 Little information 705 0.546 0.498 0 1 Sufficient information 705 0.337 0.473 0 1 Extensive information 705 0.025 0.157 0 1 Note. All variables are dummy-coded (0 = No, 1 = Yes). M = mean; SD = standard deviation. All statistics are based on N = 705 valid observations. Regarding self-reported AI knowledge, 54.6% of respondents reported little familiarity with AI, 33.7% sufficient knowledge, 2.5% extensive knowledge, and 9.0% indicated no prior information. This pattern suggests broad awareness but uneven depth of understanding – an important condition for analysing how ethical evaluations and anxiety jointly shape attitudes toward AI. Comparisons with (INSTAT, 2024 ) data (Table 2 ) indicate that the sample broadly reflects Albania’s higher-education population in terms of gender, age, and level of study, albeit with a modest overrepresentation of postgraduate students. This comparability supports the external validity of the findings and suggests that observed attitudinal patterns are likely indicative of broader tendencies among Albanian university students. In comparative terms, the study sample aligns closely with national higher-education demographics reported by INSTAT ( 2024 ). Although women are slightly overrepresented in the sample (71.4% vs. 59% nationally), this reflects the disciplinary profile of the University of Shkodra, where enrolment is concentrated in education, business, and social sciences—fields characterised by higher female participation. Table 2 Comparative profile of higher-education students in Albania, University of Shkodra (UNISHK), and study sample (2024) Characteristic National higher-education population (%)¹ UNISHK student population (%)² Study sample (%) Comment Gender Female 59 Male 41 Female ≈ 68 Male ≈ 32 Female 71.4 Male 28.6 Slight female over-representation, consistent with social-science and education dominance Level of study Bachelor 55 Master 38 Doctorate 1 Short cycle 6 Bachelor ≈ 60 Master ≈ 35 Doctorate ≈ 2 Short cycle ≈ 3 Bachelor 45 Master 52 Doctorate 3 Higher share of postgraduates, reflecting graduate research focus Type of institution Public 71 Private 29 100 Public 100 Public Sample drawn entirely from a public university Age distribution³ Majority 18–25 (≈ 65–70%) 18–25 ≈ 60 26–35 ≈ 25 36 + ≈ 15 18–25 56.9 26–35 25.5 36–50 17.6 Comparable youth concentration with inclusion of mature learners Total enrolment (N) 122 613 students ≈ 6 000 students 705 respondents – ¹ Source: Institute of Statistics (INSTAT), Studentë të regjistruar në arsimin e lartë sipas programeve të studimit, 2024. ² Approximate distribution based on internal University of Shkodra “Luigj Gurakuqi” enrolment statistics, 2024. ³ Age breakdown estimated from enrolment by cycle and typical entry age. Similarly, while bachelor students represent the majority nationally (55%), the sample includes a somewhat higher proportion of master’s students (52%), consistent with the university’s research-oriented focus and strong graduate enrolment. Age composition also mirrors national patterns, with enrolment concentrated in the 18–25 age group (56.9%), alongside meaningful representation of 26–35 (25.5%) and 36–50 (17.6%) cohorts. This generational spread enables analysis across different stages of academic and professional experience. Overall, the demographic structure of the 705 respondents closely reflects the broader Albanian higher-education population—female-dominated, youth-centred, and primarily bachelor- and master-level – while retaining sufficient postgraduate and mature-learner representation. This comparability supports the external validity of the findings and suggests that the observed ethical and emotional patterns toward AI are likely indicative of broader trends among university students in Albania. 5. Confirmatory factor analysis 5.1. Data analysis All analyses were conducted in Stata 18. Item-level descriptive statistics (mean, standard deviation, skewness, kurtosis) were inspected to assess distributional properties. Values of skewness and kurtosis within the range − 2 to + 2 were treated as acceptable for univariate normality (Kline, 2016), supporting the use of parametric procedures. To evaluate factorial validity, confirmatory factor analyses (CFA) were estimated in Stata’s SEM framework using maximum likelihood with robust standard errors (MLR). For each instrument – GAAIS, AT-EAI, and AIAS – two competing specifications were tested: (i) a one-factor model and (ii) the theoretically specified multi-factor model (two-factor for GAAIS, five-factor for AT-EAI, two-factor for AIAS). To ensure comparability across predictors, all scale scores used in regression were computed as mean indices of their respective items; AT-EAI and GAAIS remain on 1–5 metrics, while AIAS remains on a 1–7 metric, and coefficients are therefore interpreted as partial associations within each scale’s natural unit (standardised βs are reported for cross-predictor comparison). Model fit was assessed using χ²/df (acceptable range 1–3), CFI and TLI (≥ 0.90, preferably ≥ 0.95), and RMSEA and SRMR (≤ 0.08 indicating good fit). Model parsimony was compared using the AIC, where lower values indicate better fit and differences > 2 suggest meaningful improvement. Internal reliability was assessed using Cronbach’s α and McDonald’s ω, with ≥ 0.70 indicating acceptable consistency. Composite reliability (CR) and average variance extracted (AVE) were also computed to evaluate convergent validity (CR > 0.70; AVE > 0.50). Across the three scales, reliability and validity metrics met or exceeded recommended thresholds, supporting the adequacy of the Albanian versions for subsequent predictive modelling. 5.2. Findings The first study aimed to verify whether the three instruments—the General Attitudes toward Artificial Intelligence Scale (GAAIS), the Attitudes toward Ethical Artificial Intelligence Scale (AT-EAI), and the AI Anxiety Scale (AIAS)—retain their original factorial structure when applied to Albanian higher-education students. Overall, the CFAs confirmed very good model fit for all three measures, with fit indices meeting or exceeding conventional thresholds, and with factor loadings that were statistically significant and of satisfactory magnitude. General Attitudes toward AI (GAAIS) The hypothesised two-factor structure (Positive vs. Negative attitudes) fit the data very well: χ² (169) = 228.41, p = 0.003; χ²/df = 1.35; CFI = 0.987; TLI = 0.986; RMSEA = 0.031 (90% CI [0.018, 0.044]); SRMR = 0.064. All items loaded significantly (p < 0.001) on their intended factors, confirming that a unidimensional solution is unnecessary. For Positive GAAIS, standardized loadings were moderate-to-high (0.48–0.80). The strongest indicators emphasised enthusiasm and anticipated societal benefit, including “AI is exciting” (λ = 0.798), “Much of society will benefit from a future with AI” (λ = 0.769), and “I am interested in using AI systems in my daily life” (λ = 0.739). Items reflecting comparative performance claims (e.g., outperforming humans in routine jobs) loaded somewhat lower (≈ 0.52–0.53), suggesting that positivity in this sample is driven more by future-oriented optimism and engagement than by narrowly productivity-based beliefs. Internal consistency was high (α = 0.887). Table 3 presents the standardized factor loadings, estimates, and item-level descriptive statistics for the Positive and Negative GAAIS dimensions. >>>>>>>>>>>>>>>>> Table 3 here <<<<<<<<<<<<<<<<<<< Table 3 Confirmatory Factor Loadings for the Albanian GAAIS Item (indicator) Estimate SE z p 95% CI Lower 95% CI Upper Standardized Estimate Mean SD Positive GAAIS For routine transactions, I would rather interact with an AI system than a human. 0.602 0.031 19.15 < 0.001 0.540 0.663 0.482 3.512 1.142 AI can provide new economic opportunities for this country. 0.656 0.028 22.88 < 0.001 0.600 0.713 0.604 3.984 0.962 AI systems can help people feel happier. 0.529 0.025 21.08 < 0.001 0.480 0.578 0.548 3.226 0.876 I am impressed by what AI can do. 0.473 0.026 17.65 < 0.001 0.420 0.525 0.520 4.286 0.812 I am interested in using AI systems in my daily life. 0.841 0.034 24.12 < 0.001 0.773 0.910 0.739 3.567 1.053 AI can have positive impacts on people’s well-being. 0.648 0.027 23.50 < 0.001 0.594 0.702 0.714 3.874 0.861 AI is exciting. 0.711 0.029 23.87 < 0.001 0.653 0.770 0.798 4.182 0.904 An AI agent would outperform humans in many routine jobs. 0.597 0.032 18.39 < 0.001 0.534 0.661 0.526 3.224 1.156 There are many beneficial applications of AI. 0.558 0.024 22.44 < 0.001 0.509 0.607 0.682 4.351 0.744 AI systems can perform better than humans. 0.535 0.028 18.72 < 0.001 0.479 0.591 0.522 3.476 0.957 Much of society will benefit from a future with AI. 0.864 0.035 24.37 < 0.001 0.795 0.934 0.769 3.583 1.042 I would like to use AI in my own job. 0.801 0.037 21.53 < 0.001 0.728 0.874 0.661 2.978 1.231 Negative GAAIS Organizations use AI unethically. 0.428 0.0316 13.57 < 0.001 0.367 0.490 0.413 2.764 0.984 AI systems make many errors. 0.312 0.028 11.03 < 0.001 0.256 0.367 0.308 2.876 0.942 I find AI sinister. 0.942 0.041 22.872 < 0.001 0.861 1.023 0.854 3.426 1.129 AI might take control of people. 0.731 0.038 19.00 < 0.001 0.656 0.807 0.603 3.314 1.188 AI is dangerous. 0.817 0.039 20.80 < 0.001 0.740 0.894 0.694 3.012 1.166 I shiver with discomfort when thinking about AI’s future uses. 1.038 0.042 24.37 < 0.001 0.954 1.123 0.826 3.496 1.247 People like me will suffer if AI is used more and more. 0.709 0.037 18.75 < 0.001 0.635 0.783 0.566 3.152 1.175 AI is used to spy on people. 0.577 0.035 16.26 < 0.001 0.507 0.646 0.472 2.754 1.132 For Negative GAAIS, several items showed very high loadings, particularly those capturing visceral discomfort and perceived threat, such as “I find AI sinister” (λ = 0.854) and “I shiver with discomfort when thinking about AI’s future uses” (λ = 0.826). Items reflecting broader institutional scepticism (e.g., “Organizations use AI unethically”, λ = 0.413; “AI is used to spy on people”, λ = 0.472) also loaded significantly but were weaker, indicating that affective threat perceptions dominate the negative dimension more than purely institutional distrust. Reliability was good (α = 0.829). The two factors were moderately correlated (r = 0.392, p < 0.001), consistent with ambivalent technology attitudes: respondents may simultaneously recognise AI’s potential while remaining concerned about risks and loss of control. Attitudes toward Ethical AI (AT-EAI) The five-factor model of the AT-EAI also demonstrated an excellent fit to the data: χ²(413) = 502.16, p = 0.072; χ²/df = 1.22; CFI = 0.989; TLI = 0.987; RMSEA = 0.024 (90% CI [0.011, 0.037]); SRMR = 0.041. These results confirm that Albanian students are able to differentiate between distinct ethical dimensions of AI-fairness, transparency, non-maleficence, privacy, and responsibility-rather than treating ethics as a single undifferentiated concern. All items loaded significantly (p < 0.001) on their intended dimensions and most standardized loadings were comfortably above 0.70. The transparency and responsibility factors showed some of the strongest loadings (e.g. “AI developers should disclose how algorithms make decisions”, λ = 0.802; “There should be clear laws assigning responsibility for AI errors”, λ = 0.816), which suggests that, in this context, students place particular weight on explainability and accountability in AI systems. This is coherent with the institutional situation in Albania, where rules on AI use, data sharing, and algorithmic oversight are still being shaped, so students appear to demand clearer lines of responsibility. Table 4 reports the standardized factor loadings and item-level descriptive statistics for each of the five AT-EAI subscales. >>>>>>>>>>>>>>>>> Table 4 here <<<<<<<<<<<<<<<<<< Table 4 Confirmatory Factor Loadings for the Albanian AT-EAI Scale Item (indicator) Estimate SE z p 95% CI Lower 95% CI Upper Standardized Estimate Mean SD Fairness AI systems should make decisions that treat all people equally. 0.641 0.031 20.86 < 0.001 0.580 0.702 0.613 4.18 0.78 AI should be designed to avoid bias or discrimination. 0.712 0.028 25.33 < 0.001 0.657 0.767 0.744 4.26 0.73 AI applications must ensure fairness across social groups. 0.681 0.029 23.48 < 0.001 0.624 0.738 0.711 4.12 0.79 Transparency The decisions made by AI should be explainable to users. 0.728 0.030 24.27 < 0.001 0.669 0.787 0.751 4.09 0.81 AI developers should disclose how algorithms make decisions. 0.816 0.032 25.48 < 0.001 0.753 0.879 0.802 4.02 0.85 People should be informed when AI influences outcomes. 0.688 0.029 23.27 < 0.001 0.631 0.745 0.703 3.97 0.88 Transparency improves trust in AI systems. 0.762 0.031 24.72 < 0.001 0.701 0.823 0.771 4.15 0.79 Non-maleficence AI should never harm or endanger humans. 0.812 0.035 23.20 < 0.001 0.744 0.880 0.795 4.33 0.72 Safety checks must prevent harmful AI behavior. 0.755 0.033 22.73 < 0.001 0.690 0.820 0.768 4.25 0.75 AI must protect users from unintended negative effects. 0.643 0.030 21.43 < 0.001 0.584 0.702 0.675 4.17 0.78 Avoiding harm should be the primary goal of AI regulation. 0.701 0.031 22.53 < 0.001 0.640 0.762 0.712 4.20 0.77 Privacy AI systems must protect users’ personal data. 0.721 0.032 22.38 < 0.001 0.658 0.784 0.735 4.28 0.74 Data collected by AI should not be shared without consent. 0.803 0.034 23.62 < 0.001 0.737 0.869 0.781 4.36 0.71 Users should be able to control what data AI collects. 0.677 0.031 21.84 < 0.001 0.616 0.738 0.695 4.11 0.80 Responsibility Developers should be held accountable for AI outcomes. 0.724 0.030 24.13 < 0.001 0.666 0.782 0.744 4.09 0.79 There should be clear laws assigning responsibility for AI errors. 0.853 0.036 23.69 < 0.001 0.783 0.923 0.816 4.24 0.77 Users of AI systems must act responsibly. 0.672 0.028 23.77 < 0.001 0.617 0.727 0.701 4.06 0.83 The non-maleficence factor was also very robust (λ = 0.643–0.812; α = 0.83), indicating that “do no harm” remains a salient normative anchor even among young users who otherwise welcome AI. Privacy items loaded between 0.677 and 0.803, and fairness items between 0.641 and 0.712, both in acceptable-to-strong ranges. Cronbach’s α coefficients ranged from 0.74 (Fairness) to 0.83 (Non-maleficence), all above the recommended 0.70 threshold, confirming internal reliability. Inter-factor correlations were positive and moderate (e.g. r_{Fairness–Transparency} = 0.41; r_{Fairness–Responsibility} = 0.31; all p < 0.001), suggesting that students who endorse one ethical aspect of AI tend to endorse the others as well, but that these are still empirically separable constructs. This is important for Study 2, because it justifies entering the five AT-EAI subscales separately into the hierarchical regression to see which specific ethical dimension actually predicts AI attitudes. AI Anxiety Scale (AIAS) The confirmatory factor analysis supported the theorised two-factor structure of the AI Anxiety Scale (AIAS), distinguishing cognitive AI anxiety from affective AI anxiety, with excellent global model fit: χ²(76) = 112.84, p = 0.004; χ²/df = 1.48; CFI = 0.993; TLI = 0.991; RMSEA = 0.026 (90% CI [0.015, 0.036]); SRMR = 0.029. All items loaded significantly (p < 0.001) on their intended latent factors. Cognitive-anxiety items exhibited strong standardized loadings (λ = 0.705–0.784), with the strongest indicators reflecting difficulty keeping pace with rapid AI developments, such as “I feel stressed when I cannot follow AI advances” (λ = 0.784) and “Being unable to keep up with AI makes me anxious” (λ = 0.758). This pattern indicates that, for Albanian students, cognitive anxiety is driven primarily by the perceived speed and complexity of AI change rather than by interaction with specific tools. Affective-anxiety items also loaded strongly on their latent factor (λ = 0.692–0.787), particularly those capturing fears of technological dependence and labour displacement, including “I fear AI could replace human workers” (λ = 0.749), “I fear humanoid robots will take jobs” (λ = 0.772), and “I fear becoming too dependent on AI technologies” (λ = 0.787). These results suggest that emotional unease surrounding AI is closely linked to concerns about autonomy and future employment. Table 5 displays the standardized factor loadings and item-level statistics for the cognitive and affective dimensions of the Albanian AI Anxiety Scale (AIAS). >>>>>>>>>>>>>>>>> Table 5 here <<<<<<<<<<<<<<<<<< Table 5 Confirmatory Factor Loadings for the Albanian AIAS Scale Item (indicator) Estimate SE z p 95% CI Lower 95% CI Upper Standardized Estimate Mean SD AI Cognitive Anxiety Learning to use AI products makes me anxious. 0.784 0.037 21.18 < 0.001 0.711 0.857 0.768 2.91 1.03 Understanding AI functions feels overwhelming. 0.733 0.035 20.94 < 0.001 0.664 0.802 0.732 2.86 1.05 Reading an AI manual makes me nervous. 0.689 0.033 20.88 < 0.001 0.624 0.754 0.705 2.78 1.02 I feel stressed when I cannot follow AI advances. 0.811 0.038 21.29 < 0.001 0.737 0.885 0.784 2.97 1.01 AI courses would make me feel anxious. 0.753 0.034 22.12 < 0.001 0.686 0.820 0.770 2.84 1.04 Interacting with AI products makes me uneasy. 0.729 0.032 22.78 < 0.001 0.666 0.792 0.751 2.81 1.08 Learning about AI functions increases my anxiety. 0.795 0.036 22.08 < 0.001 0.725 0.865 0.776 2.95 1.06 Being unable to keep up with AI makes me anxious. 0.763 0.034 22.23 < 0.001 0.696 0.830 0.758 3.02 1.02 AI Affective Anxiety I fear AI could make me dependent. 0.712 0.033 21.64 < 0.001 0.647 0.777 0.726 3.07 1.09 I fear AI will make me lazier. 0.671 0.031 21.65 < 0.001 0.611 0.731 0.692 3.22 1.05 I fear AI could replace human workers. 0.764 0.036 21.25 < 0.001 0.693 0.835 0.749 3.34 1.07 I fear humanoid robots will take jobs. 0.801 0.037 21.67 < 0.001 0.729 0.873 0.772 3.41 1.08 I fear becoming too dependent on AI technologies. 0.828 0.039 21.23 < 0.001 0.752 0.904 0.787 3.18 1.10 I fear AI will replace people’s work. 0.689 0.034 20.21 < 0.001 0.623 0.755 0.706 3.25 1.12 Internal consistency was excellent for both subscales (Cognitive Anxiety α = 0.91; Affective Anxiety α = 0.88). The latent factors were moderately correlated (r = 0.52, p < 0.001), indicating that while cognitive and affective anxiety are related, they represent empirically distinct dimensions. This distinction is substantively important, as both dimensions were entered simultaneously into the predictive models and demonstrated differential effects on positive and negative attitudes toward AI. Taken together, the CFA results confirm that the AIAS retains its original factorial structure in the Albanian higher-education context, with strong construct validity and reliability. Alongside the GAAIS and AT-EAI, the AIAS provides a robust measure of the emotional component of AI attitudes, supporting the joint use of all three instruments in subsequent predictive analyses—the primary objective of Study 1. 6. Prediction of GAAIS by Ethical Attitudes toward AI (AT-EAI) and Anxiety (AIAS) towards AI measures This second stage addressed RQ2: To what extent do ethical attitudes toward AI (AT-EAI) and AI-related anxiety (AIAS) explain variation in students’ positive and negative attitudes toward AI (GAAIS), over and above demographic factors? We first assessed concurrent validity through zero-order associations between GAAIS and the ethical/anxiety dimensions (Table 6 ). We then estimated two hierarchical multiple regression models predicting Positive GAAIS and Negative GAAIS (Tables 7 and 8 ). This approach evaluates predictive validity, i.e., whether ethical cognition and AI anxiety explain AI attitudes in theoretically expected directions once sociodemographic heterogeneity is controlled for. 6.1. Concurrent validity: zero-order associations Table 6 shows a coherent pattern of associations between the GAAIS dimensions and the ethical (AT-EAI) and anxiety (AIAS) components. For Positive GAAIS, correlations with the five AT-EAI subscales were generally small and positive. Fairness (r = 0.043, p = 0.31) and transparency (r = 0.063, p = 0.12) were positive but non-significant, whereas non-maleficence showed a small but significant association (r = 0.117, p = 0.004). This suggests that, in this context, favourable orientations toward AI are most clearly supported by the belief that AI should avoid harm, rather than by fairness or transparency alone. >>>>>>>>>>>>>>>>> Table 6 here <<<<<<<<<<<<<<<<<< Table 6 Associations between the General Attitudes toward Artificial Intelligence Scale and the Ethical Attitudes toward AI (AT-EAI) and AI Anxiety Components (AIAS) Fairness Transparency Non-maleficence Privacy Responsibility Cognitive Anxiety Affective Anxiety Positive GAAIS r 0.043 0.063 0.117 -0.020 -0.012 -0.141 0.083 p 0.31 0.12 0.004 0.60 0.74 < 0.001 0.04 Negative GAAIS r -0.107 -0.154 -0.016 0.184 0.146 -0.407 -0.455 p 0.007 < 0.001 0.68 < 0.001 < 0.001 < 0.001 < 0.001 The anxiety components were also significantly related to positive attitudes, though with a mixed pattern: cognitive anxiety correlated negatively with Positive GAAIS (r = − 0.141, p < 0.001), while affective anxiety correlated positively but weakly (r = 0.083, p = 0.04). This indicates that students who feel overwhelmed by AI developments tend to be less optimistic, whereas mild affective concern does not necessarily eliminate enthusiasm for AI’s potential benefits. For Negative GAAIS, associations were stronger and more systematic. Ethical dimensions linked to safeguards and rights—especially privacy (r = 0.184, p < 0.001) and responsibility (r = 0.146, p < 0.001)—were positively related to negative attitudes, indicating that students who place greater emphasis on data protection and accountability are also more alert to AI risks. Conversely, fairness (r = − 0.107, p = 0.007) and transparency (r = − 0.154, p < 0.001) were negatively associated with negative attitudes, suggesting that believing AI can be designed to be fair and explainable corresponds to lower levels of AI “demonisation.” The strongest concurrent associations, however, were observed for anxiety: both cognitive anxiety (r = − 0.407, p < 0.001) and affective anxiety (r = − 0.455, p < 0.001) showed large associations with negative attitudes, consistent with the view that anxiety functions as a primary affective driver of perceived AI threat. Overall, the correlation matrix supports concurrent validity: the GAAIS dimensions relate to ethical cognition and anxiety in theoretically meaningful ways, but ethical dimensions are not equally salient, and anxiety exhibits the largest associations – especially with negative AI attitudes. 6.2. Predicting positive GAAIS: the role of non-maleficence and anxiety Table 7 presents the full hierarchical model predicting Positive GAAIS, which explains a substantial share of variance (R² = 0.472; F(20, 684) = 7.08; p < 0.001). In Block 1 (demographics), two results stand out. First, male students reported significantly more positive attitudes than female students (B = 0.215, β = 0.11, p = 0.016). Second, self-reported AI information was negatively associated with Positive GAAIS: compared with those reporting “no information,” students reporting “little” (B = − 0.293, p = 0.028), “enough” (B = − 0.568, p < 0.001), and especially “much” AI information (B = − 1.127, p >>>>>>>>>>>>>>>> Table 7 here <<<<<<<<<<<<<<<<<< Table 7 Coefficients for the Positive GAAIS full model (Model 3) B SE β t p 95% CI B lower 95% CI B upper (Constant) 0.645 0.296 - 2.18 0.030 0.064 1.227 Demographics Male 0.215 0.090 0.11 2.41 0.016 0.040 0.391 Age 18–25 0.096 0.130 0.04 0.74 0.461 −0.159 0.351 Age 26–35 −0.069 0.117 −0.03 −0.58 0.559 −0.299 0.162 Bachelor student −0.394 0.236 −0.12 −1.67 0.095 −0.857 0.069 Bachelor degree −0.569 0.235 −0.17 −2.42 0.016 −1.030 −0.108 Master student −0.471 0.241 −0.14 −1.96 0.051 −0.943 0.002 Master degree −0.361 0.219 −0.10 −1.65 0.100 −0.791 0.069 Little AI information −0.293 0.133 −0.09 −2.20 0.028 −0.555 −0.031 Enough AI information −0.568 0.141 −0.17 −4.02 0.000 −0.846 −0.291 Much AI information −1.127 0.265 −0.21 −4.25 0.000 −1.648 −0.606 AT-EAI dimensions Fairness 0.017 0.037 0.02 0.45 0.651 −0.056 0.090 Transparency 0.055 0.040 0.06 1.38 0.169 −0.024 0.134 Non-maleficence 0.125 0.036 0.17 3.45 0.001 0.054 0.196 Privacy −0.013 0.036 −0.01 −0.37 0.711 −0.084 0.058 Responsibility 0.019 0.039 0.02 0.49 0.625 −0.057 0.095 AI Anxiety components Cognitive anxiety −0.164 0.025 −0.25 −6.56 < 0.001 −0.213 −0.115 Affective anxiety −0.191 0.026 −0.29 −7.35 < 0.001 −0.243 −0.139 Although counter-intuitive, this pattern plausibly reflects that greater exposure is accompanied by greater awareness of limitations, risks, and contested uses (e.g., generative AI in assessment contexts, proctoring, or plagiarism detection), thereby reducing uncritical enthusiasm. In Block 2 (ethical attitudes), only non-maleficence significantly predicted higher Positive GAAIS (B = 0.125, β = 0.17, p = 0.001). Fairness, transparency, privacy, and responsibility were non-significant. This reinforces the bivariate pattern: safety and harm-avoidance are the ethical dimensions most strongly tied to favourable AI orientations in this sample. In Block 3 (AI anxiety), both anxiety dimensions emerged as strong negative predictors: cognitive anxiety (B = − 0.164, β = −0.25, p < 0.001) and affective anxiety (B = − 0.191, β = −0.29, p < 0.001). Holding demographics and ethical cognition constant, anxiety appears to be the most powerful suppressor of positive AI attitudes. Students who feel unable to keep pace with AI developments or who fear dependence and displacement report markedly lower optimism, even when they endorse ethical AI principles—highlighting the centrality of the affective pathway in attitude formation. 6.3. Predicting negative GAAIS: anxiety and privacy as drivers of AI risk salience Table 8 presents the parallel model predicting Negative GAAIS, which is also statistically significant (R² = 0.392; F(20, 684) = 6.88; p >>>>>>>>>>>>>>>> Table 8 here <<<<<<<<<<<<<<<<<< Table 8 Coefficients for the Negative GAAIS full model (Model 3) B SE β t p 95% CI B lower 95% CI B upper (Constant) 0.041 0.296 — 0.14 0.890 −0.541 0.622 Demographics Male (1 = male) 0.332 0.090 0.17 3.70 0.000 0.156 0.507 Age 18–25 (ref = 36–50) 0.215 0.130 0.10 1.65 0.099 −0.040 0.471 Age 26–35 0.227 0.117 0.11 1.93 0.054 −0.004 0.457 Bachelor student −0.383 0.236 −0.12 −1.62 0.105 −0.847 0.080 Bachelor degree −0.398 0.235 −0.12 −1.69 0.091 −0.859 0.064 Master student −0.196 0.241 −0.06 −0.81 0.417 −0.668 0.277 Master degree −0.136 0.219 −0.04 −0.62 0.534 −0.567 0.294 Little AI information −0.131 0.133 −0.04 −0.98 0.327 −0.393 0.131 Enough AI information −0.231 0.141 −0.07 −1.64 0.102 −0.509 0.046 Much AI information −0.028 0.266 −0.01 −0.10 0.917 −0.549 0.494 AT-EAI dimensions Fairness −0.053 0.037 −0.06 −1.41 0.159 −0.126 0.021 Transparency −0.074 0.040 −0.08 −1.84 0.066 −0.153 0.005 Non-maleficence −0.051 0.036 −0.06 −1.40 0.161 −0.122 0.020 Privacy 0.092 0.036 0.10 2.55 0.011 0.021 0.163 Responsibility 0.037 0.039 0.04 0.95 0.343 −0.040 0.113 AI Anxiety components Cognitive anxiety 0.151 0.030 0.29 5.03 < 0.001 0.092 0.210 Affective anxiety 0.167 0.031 0.32 5.38 < 0.001 0.106 0.228 Among demographic controls, male students again scored higher (B = 0.332, β = 0.17, p < 0.001). Taken together with the positive-attitude findings, this indicates a more polarised or salient attitudinal profile among male students – greater endorsement of both promise and risk – rather than a simple optimism–pessimism shift. Age and education were not statistically significant in the full model. Among ethical attitudes, only privacy predicted higher negative attitudes (B = 0.092, β = 0.10, p = 0.011). Students who place greater importance on consent, data protection, and control over personal information are more likely to endorse risk-oriented statements. Fairness and non-maleficence had negative coefficients, as expected, but were not significant once anxiety was included, suggesting that ethical reassurance may partly operate through emotional responses or be overshadowed by affective threat perceptions. As hypothesised, AI anxiety was the dominant predictor of negative GAAIS. Both cognitive anxiety (B = 0.151, β = 0.29, p < 0.001) and affective anxiety (B = 0.167, β = 0.32, p < 0.001) were strong and highly significant. Standardised coefficients indicate that affective anxiety was the single strongest predictor (β = 0.32), slightly exceeding cognitive anxiety (β = 0.29), implying that fear-based concerns (dependency and displacement) are especially influential in elevating perceived AI risk. 6.4. Concurrent and predictive validity Together, results provide strong evidence of both concurrent and predictive validity for the Albanian adaptations of the AT-EAI, AIAS, and GAAIS. Ethical and affective predictors jointly explain substantial variance in attitudes after controlling for demographic differences (R² = 0.47 for Positive GAAIS; R² = 0.39 for Negative GAAIS). Across models, two patterns are especially robust. First, among ethical dimensions, non-maleficence is the key facilitator of positive attitudes, while privacy is the key amplifier of negative attitudes—suggesting that safety and data protection are the most consequential ethical “entry points” shaping students’ evaluations of AI in higher education. Second, both forms of AI anxiety are consistently strong predictors: cognitive anxiety and affective anxiety suppress positive attitudes and intensify negative attitudes, with affective anxiety showing the largest effect on risk salience. Gender differences are noteworthy: male students reported both stronger positive and stronger negative attitudes, consistent with a more polarised attitude profile. Finally, greater self-reported AI information predicted lower positive attitudes, suggesting that increased familiarity may be accompanied by more critical awareness of limitations and contested uses of AI tools in academic settings. Although counter-intuitive, this pattern is consistent with an ‘informed scepticism’ mechanism: greater exposure may increase awareness of contested uses (e.g., surveillance-like proctoring, opaque plagiarism detection, dataset bias, and academic-integrity dilemmas), reducing uncritical enthusiasm. In contexts where institutional guidance is fragmented, information may be acquired through social media narratives and peer experience, which can amplify critical frames rather than build calibrated trust. Table 9 summarizes the research hypotheses alongside the observed results from the predictive models, highlighting areas of support and divergence. >>>>>>>>>>>>>>>>> Table 9 here <<<<<<<<<<<<<<<<<< Table 9 Summary of Hypotheses and Results Predictor Hypothesized Direction Positive GAAIS (Observed) Hypothesized Direction Negative GAAIS (Observed) Fairness Positive n.s. Positive n.s. Transparency Positive n.s. Positive Negative (marginal, p ≈ .07) Non-maleficence Positive Positive (p < .01) Positive n.s. Privacy Positive n.s. Positive Positive (p < .05) Responsibility Positive n.s. Positive n.s. AI Anxiety (Cognitive) Negative Negative (p < .001) Negative Positive (p < .001) → higher anxiety → greater risk salience AI Anxiety (Affective) Negative Negative (p < .001) (weak positive bivariate p < .05) Negative Positive (p < .001) (strongest predictor) Gender (1 = Male) n/a Positive (p < .05) n/a Positive (p < .001) Age n/a n.s. n/a n.s. Education level Positive Negative (p < .10) Positive n.s. AI Information level Positive Negative (p < .001) Positive n.s. From a policy and practice perspective, the findings imply that ethical reassurance alone is insufficient to foster favourable attitudes unless anxiety is simultaneously addressed and visible safeguards are in place. For Albanian higher education, AI literacy efforts may be most effective when they integrate technical competence, ethical reflection, and affective preparedness, with particular emphasis on “do no harm” principles and data protection by design in AI-enabled teaching and assessment systems. 7. Discussion This study examined how ethical cognition and emotional responses jointly shape university students’ general attitudes toward artificial intelligence (AI) in Albania. Drawing on validated Albanian versions of the GAAIS, AT-EAI, and AIAS, the findings show that students’ evaluations of AI are structurally ambivalent: optimism about AI’s potential coexists with salient apprehension. This pattern is consistent with contemporary views of technology attitudes as multi-dimensional rather than a single acceptance–rejection continuum. The results support affective–cognitive models of attitude formation (Ajzen & Fishbein; Eagly & Chaiken) by demonstrating that ethical evaluations and emotional reactions are empirically distinguishable and jointly relevant. Importantly, not all ethical concerns mattered equally. Non-maleficence (“do no harm”) was the only ethical dimension that reliably supported more positive attitudes, suggesting that safety functions as the primary moral “threshold condition” for endorsing AI in higher education. By contrast, privacy concerns amplified negative attitudes, indicating that data protection and control over personal information act as a rights-based lens through which AI risks become more salient—particularly in educational environments where monitoring, assessment, and data processing are central. AI anxiety emerged as the strongest and most consistent predictor across outcomes. Both cognitive anxiety (difficulty understanding or keeping up with AI) and affective anxiety (fear of dependence or displacement) reduced positive attitudes and increased negative attitudes, with affective anxiety showing the largest association with risk-oriented evaluations. This aligns with affect heuristic and dual-process accounts of risk perception (Slovic et al.), where emotional cues operate as rapid signals of threat under uncertainty. The moderate association between positive and negative GAAIS dimensions further indicates that endorsement of AI’s benefits does not preclude discomfort about its consequences; rather, students appear to evaluate AI through parallel “opportunity” and “risk” channels. These findings also extend technology acceptance perspectives (TAM/UTAUT2) by showing that instrumental familiarity is not sufficient to explain AI attitudes in higher education. Once ethical and affective factors are considered, demographics play a comparatively smaller role, and greater self-reported AI knowledge is associated with lower positivity. A plausible interpretation is that increased exposure brings more critical awareness of contested academic uses (e.g., integrity systems, surveillance-like monitoring, opaque decisions), reducing unreflective enthusiasm. In this sense, AI literacy may produce “informed scepticism” as well as competence, suggesting that the knowledge–acceptance relationship is not necessarily linear. The Albanian context helps interpret why non-maleficence and privacy are especially salient. Students’ encounters with AI often occur through externally introduced systems—plagiarism detection, automated evaluation, and generative AI tools—within an institutional setting where governance standards and ethical guidance are still consolidating. Under such conditions, safety and data protection become practical proxies for trust: when formal assurances are limited, students rely more strongly on moral safeguards and on their emotional sense of control. From a higher-education perspective, the results imply that responsible AI integration should not be framed as a purely technical upgrade. Universities are likely to see better acceptance when AI literacy is paired with (i) visible harm-avoidance safeguards and (ii) clear data-governance practices that make privacy protections concrete. Equally, anxiety is not a peripheral attitude “noise” but a core mechanism shaping receptiveness; reducing anxiety requires structured opportunities for guided exposure, critical discussion, and transparent explanation of how AI-enabled systems work and how students are protected. Several limitations should be noted. The cross-sectional design limits causal inference, and reliance on self-report measures may inflate associations through shared method variance. Future work should use longitudinal or experimental designs (e.g., pre/post AI-ethics training) to assess whether anxiety decreases and attitudes shift with structured exposure and clearer institutional safeguards. Extending the model to include perceived control, institutional trust, and perceived fairness of university AI practices would also help clarify the pathways through which ethics and anxiety translate into positive and negative evaluations. This study offers compelling evidence that attitudes toward artificial intelligence in Albanian higher education are shaped by an interplay of ethical cognition and affective responses. The validated Albanian versions of the AT-EAI, AIAS, and GAAIS demonstrated excellent psychometric properties, enabling a rigorous examination of how moral evaluations—particularly non-maleficence and privacy—and emotional factors jointly predict both positive and negative AI attitudes. Ethical endorsement of non-maleficence emerged as the most consistent driver of favourable evaluations, whereas privacy concerns heightened sensitivity to perceived risks. Affective and cognitive facets of anxiety proved especially consequential: both significantly suppressed positive attitudes and intensified negative ones, with affective anxiety exerting the strongest influence on risk perception. These findings reinforce the centrality of emotion within technology-acceptance processes and point to broader dynamics characteristic of post-transition institutional contexts, where ethical reassurance often compensates for limited regulatory clarity or institutional trust. Theoretically, the study integrates cognitive, affective, and ethical dimensions into a unified framework of AI acceptance, demonstrating that attitudes toward AI extend beyond utilitarian appraisals to encompass deeper moral and emotional layers. Despite its contributions, the study’s cross-sectional design and reliance on self-report measures constrain causal inference and raise the possibility of common-method bias. Future work should therefore adopt longitudinal, experimental, and cross-cultural designs to further refine and generalize the proposed model. To strengthen students’ engagement with artificial intelligence, higher education institutions should integrate both ethical and emotional dimensions into AI literacy programs, ensuring that learners develop critical understanding and reduced anxiety. Institutional trust must also be reinforced through transparent data policies, clear accountability mechanisms, and human oversight, which are essential in transitional governance contexts. Universities should implement targeted, evidence-based interventions—such as workshops, reflective discussions, and counselling services—to address cognitive and affective dimensions of AI-related anxiety. Future research should examine how contextual and cultural factors shape AI perceptions, with particular attention to the role of institutional trust. Finally, adopting longitudinal and experimental research designs would improve causal inference and track how ethical cognition and anxiety evolve over time. Declarations Ethics approval and consent to participate This study was conducted in accordance with the ethical standards of the University of Shkodra “Luigj Gurakuqi” and with the principles of the Declaration of Helsinki. Ethical approval was obtained from the University Ethics Committee of the University of Shkodra “Luigj Gurakuqi” (Approval No. 2025/AEI-05). All participants were informed about the purpose of the study, the voluntary nature of participation, and their right to withdraw at any time without consequence. Written informed consent was obtained from all participants prior to data collection. Consent for publication Not applicable. The manuscript does not contain any individual person’s data in any identifiable form. Availability of data and materials The datasets generated and/or analysed during the current study are not publicly available due to ethical and privacy considerations related to the protection of participants’ anonymity, but are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Authors’ contributions E.D. conceived the study, designed the research framework, performed the statistical analyses, and drafted the manuscript. E.H. translated and culturally adapted the instruments, conducted pilot testing, coordinated data collection, and contributed to data preparation and validation. Both authors reviewed, revised, and approved the final manuscript. Acknowledgements The authors wish to thank the students and academic staff of the University of Shkodra “Luigj Gurakuqi” for their participation and valuable insights. We are also grateful to the University Ethics Committee for their guidance and approval of the research protocol. Special thanks to colleagues from the Faculty of Economy and the Faculty of Educational Sciences for their feedback during the instrument adaptation and piloting phases. Finally, we acknowledge the support of the university’s administrative and IT staff in facilitating both online and in-class data collection. Authors’ information E.D. is a lecturer in research methods and applied economics at the department of Business Administration, Faculty of Economy, University of Shkodra “Luigj Gurakuqi”, with research interests in labour markets, inequality, technology acceptance, and the social and ethical implications of artificial intelligence. E.H. is a lecturer in psychology and serves as the head of the Social Work and Psychology Department, Faculty of Educational Sciences, the University of Shkodra “Luigj Gurakuqi”, with research interests in educational psychology, measurement, and digital transformation in education. References Ajzen, I., 1982. Equity in Attitude Formation and Change, in: Greenberg, J., Cohen, R.L. (Eds.), Equity and Justice in Social Behavior. Academic Press, pp. 161–186. https://doi.org/10.1016/B978-0-12-299580-4.50011-2 Ajzen, I., Fishbein, M., 2000. Attitudes and the Attitude-Behavior Relation: Reasoned and Automatic Processes. European Review of Social Psychology 11, 1–33. https://doi.org/10.1080/14792779943000116 Arora, N., Garg, N., 2023. Meaningful Work in the Digital Age- A Comprehensive Review and Framework. Human Resource Development International 0, 1–25. https://doi.org/10.1080/13678868.2024.2336866 Bankins, S., Formosa, P., 2023. The Ethical Implications of Artificial Intelligence (AI) For Meaningful Work. J Bus Ethics 185, 725–740. https://doi.org/10.1007/s10551-023-05339-7 Cicero, L., Russo, A., Di Stefano, G., Zammitti, A., 2025. The General Attitudes towards Artificial Intelligence Scale (GAAIS): validation and psychometric properties analysis in the Italian context. BMC Psychol 13, 641. https://doi.org/10.1186/s40359-025-02935-2 Council of Ministers, 2025. Albania 2030 – Technology as a Pillar of Integration into the European Union. Albanian Government Council of Ministers. URL https://www.kryeministria.al/en/newsroom/shqiperia-2030-teknologjia-si-shtylle-e-integrimit-ne-bashkimin-europian/ (accessed 12.21.25). Davis, F.D., 1989. Technology acceptance model: TAM. Al-Suqri, MN, Al-Aufi, AS: Information Seeking Behavior and Technology Adoption 205. Eagly, A.H., Chaiken, S., 1993. The psychology of attitudes, The psychology of attitudes. Harcourt Brace Jovanovich College Publishers, Orlando, FL, US. ETF, 2022. Embracing the digital age | ETF. EU4Youth, 2024. Youth Guarantee in Albania: Real opportunities for employment, skills and education. | UNICEF Albania. URL https://www.unicef.org/albania/stories/youth-guarantee-albania-real-opportunities-employment-skills-and-education (accessed 7.16.25). European Commission, 2025a. 2025 Communication on EU enlargement policy. European Commission, 2025b. The Commission publishes guidelines on AI system definition to facilitate the first AI Act’s rules application | Shaping Europe’s digital future. GoA, 2025. Per miratimin e Strategjise Kombetare te Inteligjences Artificiale te Shqiperise 2025-2030 dhe te Planit te Veprimit. Council of Ministers, Tirana, Albania. Hubertz, M., Janowsky, A., 2025. Mind the Gap: Understanding Diverse Student Experiences with AI Through GAAIS Analysis, in: Smith, B.K., Borge, M. (Eds.), Learning and Collaboration Technologies. Springer Nature Switzerland, Cham, pp. 252–269. https://doi.org/10.1007/978-3-031-93746-0_18 INSTAT, 2024. Arsimi - Shifrat. Jang, Y., Choi, S., Kim, H., 2022. Development and validation of an instrument to measure undergraduate students’ attitudes toward the ethics of artificial intelligence (AT-EAI) and analysis of its difference by gender and experience of AI education. Educ Inf Technol 27, 11635–11667. https://doi.org/10.1007/s10639-022-11086-5 Jobin, A., Ienca, M., Vayena, E., 2019. The global landscape of AI ethics guidelines. Nat Mach Intell 1, 389–399. https://doi.org/10.1038/s42256-019-0088-2 Kaya, F., Aydin, F., Schepman, A., Rodway, P., Yetişensoy, O., Demir Kaya, M., 2024. The Roles of Personality Traits, AI Anxiety, and Demographic Factors in Attitudes toward Artificial Intelligence. International Journal of Human–Computer Interaction 40, 497–514. https://doi.org/10.1080/10447318.2022.2151730 Kong, S.-C., Cheung, W.M.-Y., Tsang, O., 2023. Evaluating an artificial intelligence literacy programme for empowering and developing concepts, literacy and ethical awareness in senior secondary students. Educ Inf Technol 28, 4703–4724. https://doi.org/10.1007/s10639-022-11408-7 Lee, M.K., Kusbit, D., Metsky, E., Dabbish, L., 2015. Working with Machines: The Impact of Algorithmic and Data-Driven Management on Human Workers, in: Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems. Presented at the CHI ’15: CHI Conference on Human Factors in Computing Systems, ACM, Seoul Republic of Korea, pp. 1603–1612. https://doi.org/10.1145/2702123.2702548 Lim, T., Gottipati, S., Cheong, M., 2025. What students really think: unpacking AI ethics in educational assessments through a triadic framework. Int J Educ Technol High Educ 22, 56. https://doi.org/10.1186/s41239-025-00556-8 Liska, A.E., 1984. A Critical Examination of the Causal Structure of the Fishbein/Ajzen Attitude-Behavior Model. Social Psychology Quarterly 47, 61–74. https://doi.org/10.2307/3033889 Longoni, C., Cian, L., 2022. Artificial Intelligence in Utilitarian vs. Hedonic Contexts: The “Word-of-Machine” Effect. Journal of Marketing 86, 91–108. https://doi.org/10.1177/0022242920957347 Mantello, P.A., Ghotbi, N., Ho, M.-T., Mizutani, F., 2025. Gauging public opinion of AI and emotionalized AI in healthcare: findings from a nationwide survey in Japan. AI & Soc 40, 3735–3749. https://doi.org/10.1007/s00146-024-02126-4 MAS, 2025. Dokumenti i Politikave te Arsimit te Larte. Ministria e Arsimit dhe Sportit. MASR, 2021. Draft-Strategjia Kombëtare e Arsimit 2021-2026. Ministria e Arsimit, Sportit dhe Rinisë. Mubashir, A.S., Altaf, W., Kainaat, R., 2025. Personality Traits and Attitudes Towards Artificial Intelligence Among University Students. Social Science Review Archives 3, 1010–1019. https://doi.org/10.70670/sra.v3i2.690 Nomura, T., Tanaka, M., 2022. Experiences, knowledge of functions, and social acceptance of robots: an exploratory case study focusing on Japan. AI & Soc 37, 367–374. https://doi.org/10.1007/s00146-021-01196-y Sacco, F., Manzi, F., Di Dio, C., Schepman, A., Rodway, P., Massaro, D., Marchetti, A., 2025. Validation of the general attitudes towards Artificial Intelligence Scale in the Italian context and the role of trust. Computers in Human Behavior Reports 19, 100751. https://doi.org/10.1016/j.chbr.2025.100751 Schepman, A., Rodway, P., 2025. The Measurement of Attitudes Towards Artificial Intelligence: An Overview and Recommendations, in: Montag, C., Ali, R. (Eds.), The Impact of Artificial Intelligence on Societies: Understanding Attitude Formation Towards AI. Springer Nature Switzerland, Cham, pp. 9–24. https://doi.org/10.1007/978-3-031-70355-3_2 Schepman, A., Rodway, P., 2020. Initial validation of the general attitudes towards Artificial Intelligence Scale. Computers in Human Behavior Reports 1, 100014. https://doi.org/10.1016/j.chbr.2020.100014 Skagerlund, K., Forsblad, M., Slovic, P., Västfjäll, D., 2020. The Affect Heuristic and Risk Perception – Stability Across Elicitation Methods and Individual Cognitive Abilities. Front. Psychol. 11. https://doi.org/10.3389/fpsyg.2020.00970 Slovic, P., Finucane, M.L., Peters, E., MacGregor, D.G., 2007. The affect heuristic. European Journal of Operational Research 177, 1333–1352. https://doi.org/10.1016/j.ejor.2005.04.006 Slovic, P., Finucane, M.L., Peters, E., MacGregor, D.G., 2004. Risk as analysis and risk as feelings: some thoughts about affect, reason, risk, and rationality. Risk Anal 24, 311–322. https://doi.org/10.1111/j.0272-4332.2004.00433.x Sultana, A., Abdul Latheef, N., Siby, N., Ahmad, Z., 2025. Exploring Students’ Attitudes Toward Artificial Intelligence (AI): Psychometric Validation of AI-Attitude Scale. Sage Open 15, 21582440251378375. https://doi.org/10.1177/21582440251378375 Sun, L., Zhou, L., 2025. Generative artificial intelligence attitude analysis of undergraduate students and their precise improvement strategies: A differential analysis of multifactorial influences. Educ Inf Technol 30, 10591–10626. https://doi.org/10.1007/s10639-024-13236-3 Venkatesh, V., Morris, M.G., Davis, G.B., Davis, F.D., 2003. User Acceptance of Information Technology: Toward a Unified View. MIS Quarterly 27, 425–478. https://doi.org/10.2307/30036540 Wang, W., Downey, J., Yang, F., 2025. AI anxiety? Comparing the sociotechnical imaginaries of artificial intelligence in UK, Chinese and Indian newspapers. Global Media and China 10, 425–441. https://doi.org/10.1177/20594364231196547 Wang, Y.-Y., Wang, Y.-S., 2022. Development and validation of an artificial intelligence anxiety scale: an initial application in predicting motivated learning behavior. Interactive Learning Environments 30, 619–634. https://doi.org/10.1080/10494820.2019.1674887 Weng, X., Xia, Q., Ahmad, Z., Chiu, T.K.F., 2024. Personality traits for self-regulated learning with generative artificial intelligence: The case of ChatGPT. Computers and Education: Artificial Intelligence 7, 100315. https://doi.org/10.1016/j.caeai.2024.100315 Xue, L., Rashid, A.M., Ouyang, S., 2024. The Unified Theory of Acceptance and Use of Technology (UTAUT) in Higher Education: A Systematic Review. Sage Open 14, 21582440241229570. https://doi.org/10.1177/21582440241229570 Yilmaz, F.G.K., Yilmaz, R., Ceylan, M., 2024. Generative Artificial Intelligence Acceptance Scale: A Validity and Reliability Study. International Journal of Human–Computer Interaction 40, 8703–8715. https://doi.org/10.1080/10447318.2023.2288730 Zhu, X., Gan, J., Yang, X., 2025. Social mobility beliefs moderate links between SSS, attributions for wealth and poverty, and aggression. BMC Psychol 13, 820. https://doi.org/10.1186/s40359-025-03138-5 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 19 Feb, 2026 Reviews received at journal 02 Feb, 2026 Reviews received at journal 02 Feb, 2026 Reviewers agreed at journal 29 Jan, 2026 Reviewers agreed at journal 29 Jan, 2026 Reviewers invited by journal 29 Jan, 2026 Editor invited by journal 07 Jan, 2026 Editor assigned by journal 01 Jan, 2026 Submission checks completed at journal 01 Jan, 2026 First submitted to journal 30 Dec, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8483126","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":568141149,"identity":"91f0060b-a36e-4043-b449-79e645fad80b","order_by":0,"name":"Elona Hasmujaj","email":"","orcid":"","institution":"University of Shkodra \"Luigj Gurakuqi\"","correspondingAuthor":false,"prefix":"","firstName":"Elona","middleName":"","lastName":"Hasmujaj","suffix":""},{"id":568141152,"identity":"cc0776c4-e018-4994-ac9a-efb6b8920391","order_by":1,"name":"Elvisa Drishti","email":"data:image/png;base64,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","orcid":"","institution":"University of Shkodra \"Luigj Gurakuqi\"","correspondingAuthor":true,"prefix":"","firstName":"Elvisa","middleName":"","lastName":"Drishti","suffix":""}],"badges":[],"createdAt":"2025-12-30 16:38:39","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8483126/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8483126/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":99505179,"identity":"eee46c0c-139b-486d-a51a-5457225a0df2","added_by":"auto","created_at":"2026-01-05 08:27:43","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":161529,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscriptwithauthordetails1Jan2026.docx","url":"https://assets-eu.researchsquare.com/files/rs-8483126/v1/ec9f06466921ab485ff55a55.docx"},{"id":99791306,"identity":"7945166a-c53a-46a3-ba3b-31061b25773c","added_by":"auto","created_at":"2026-01-08 12:59:25","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6382,"visible":true,"origin":"","legend":"","description":"","filename":"59fa3a584c164de1b72679e4066e83cf.json","url":"https://assets-eu.researchsquare.com/files/rs-8483126/v1/e64f910dc7666bdabcaa770e.json"},{"id":99505181,"identity":"dc973190-fdba-45a0-9520-63bfc0b690df","added_by":"auto","created_at":"2026-01-05 08:27:43","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":253598,"visible":true,"origin":"","legend":"","description":"","filename":"59fa3a584c164de1b72679e4066e83cf1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8483126/v1/a1db314fc291c7dfa7e4c9e2.xml"},{"id":99791514,"identity":"d751adf1-a911-47b4-a868-f91526124d86","added_by":"auto","created_at":"2026-01-08 13:00:56","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":248906,"visible":true,"origin":"","legend":"","description":"","filename":"59fa3a584c164de1b72679e4066e83cf1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8483126/v1/049d3256ffb992e6942453fc.xml"},{"id":99505182,"identity":"fa9e7eb9-f2bb-4d9b-8869-b9114c28ca78","added_by":"auto","created_at":"2026-01-05 08:27:43","extension":"html","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":260890,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8483126/v1/14227aae43a2838e2d9e0e6e.html"},{"id":99803113,"identity":"f7b8392d-e386-4184-a5d1-e3d698a5c069","added_by":"auto","created_at":"2026-01-08 14:09:40","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1895921,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8483126/v1/66651c8e-8f17-43e0-a72c-1612f322164a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Ethical Cognition, Anxiety, and Attitudes toward Artificial Intelligence in Higher Education: Validation and Predictive Modelling of the Albanian GAAIS","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe rapid diffusion of artificial intelligence (AI) across educational, professional, and social domains has intensified scholarly interest in how individuals perceive, evaluate, and emotionally respond to intelligent systems (Bankins and Formosa, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Schepman and Rodway, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Sultana et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Within higher education, AI tools are increasingly embedded in teaching, assessment, and administrative processes, reshaping how knowledge is produced, delivered, and governed (Hubertz and Janowsky, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Mubashir et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Weng et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). While these developments promise gains in efficiency, personalisation, and accessibility, they simultaneously raise concerns related to privacy, fairness, accountability, and the preservation of human agency.\u003c/p\u003e \u003cp\u003eTo capture the multidimensional nature of these responses, the present study adopts an integrated cognitive\u0026ndash;affective\u0026ndash;ethical framework for analysing attitudes toward AI in higher education. This framework draws on three complementary theoretical traditions. First, the Attitude\u0026ndash;Behaviour Model Model (Ajzen, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1982\u003c/span\u003e; Ajzen and Fishbein, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Liska, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e1984\u003c/span\u003e) conceptualises attitudes as composite evaluations arising from the interaction between cognitive appraisals (beliefs, judgments, evaluations) and affective reactions (emotions such as anxiety, curiosity, or enthusiasm). In this study, ethical cognition represents the cognitive dimension\u0026mdash;individuals\u0026rsquo; evaluations of fairness, transparency, non-maleficence, privacy, and responsibility in AI systems\u0026mdash;while AI anxiety captures the affective dimension, encompassing both cognitive concerns (e.g., difficulty understanding AI) and emotional responses (e.g., fear of dependence or displacement) (Wang and Wang, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eSecond, insights from the Affect Heuristic and dual-process theories of risk perception (Skagerlund et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Slovic et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2004\u003c/span\u003e) emphasise that emotions are not subordinate to reason but function as parallel processing channels that strongly influence judgments under conditions of complexity and uncertainty. In the context of AI, affective cues\u0026mdash;such as fear, fascination, or unease\u0026mdash;often shape perceptions of risk, trust, and legitimacy more powerfully than objective knowledge. Ethical reassurance (cognitive) and emotional anxiety (affective) therefore operate jointly in shaping overall evaluations of AI\u0026rsquo;s desirability and social acceptability.\u003c/p\u003e \u003cp\u003eThird, the study extends insights from the Technology Acceptance Model (TAM) (Davis, 1989) and the Unified Theory of Acceptance and Use of Technology (UTAUT/UTAUT2) (Venkatesh et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), which traditionally link perceived usefulness and ease of use to technology adoption. Recent extensions of these models increasingly recognise that moral trust, perceived fairness, and emotional comfort play a central moderating role in technology acceptance, particularly for high-stakes and opaque technologies such as AI (Mubashir et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Schepman and Rodway, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Sultana et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBy integrating ethical cognition (measured through AT-EAI) and emotional response (measured through AIAS) as antecedents of general attitudes toward AI (measured through GAAIS), the present study situates AI acceptance within a broader socio-psychological and ethical framework rather than a narrowly utilitarian one. Growing empirical evidence supports this synthesis, showing that individuals\u0026rsquo; attitudes toward AI are shaped as much by moral and emotional considerations as by instrumental evaluations of performance or (Lee et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Longoni and Cian, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Ethical judgments concerning fairness, transparency, and responsibility constitute a cognitive foundation for trust, while emotional responses such as anxiety or excitement represent the affective component of technological appraisal. Together, these dimensions determine whether AI is perceived as empowering or threatening.\u003c/p\u003e \u003cp\u003eFrom an educational perspective, this inquiry is particularly salient in transitional contexts such as Albania, where digital transformation has advanced more rapidly than institutional regulation and ethical governance frameworks (Council of Ministers, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; GoA, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; MASR, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Students and academics are increasingly exposed to AI-enabled learning management systems, plagiarism detection tools, and generative AI applications, often in the absence of systematic ethical guidance or robust data governance mechanisms (MAS, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). In such environments, perceptions of AI are shaped less by formal regulation and more by informal experiences, cultural narratives, and emotional reactions. Examining how ethical awareness and AI-related anxiety jointly influence attitudes toward AI in higher education therefore offers critical insights into digital readiness, ethical literacy, and the psychology of technology acceptance in emerging economies (EU4Youth, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; GoA, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo address these issues empirically, the study employs three internationally recognised instruments. The Attitudes toward Ethical Artificial Intelligence Scale (AT-EAI) (Jang et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) assesses individuals\u0026rsquo; endorsement of ethical principles guiding AI design and use. The AI Anxiety Scale (AIAS) (Wang and Wang, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) captures cognitive and affective anxiety elicited by AI technologies. The General Attitudes toward Artificial Intelligence Scale (GAAIS) (Schepman and Rodway, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) measures both optimistic and sceptical orientations toward AI. Together, these instruments operationalise the cognitive, affective, and global components of attitude formation, providing an integrated framework for analysing how ethical cognition and emotional responses shape attitudes toward AI in higher education.\u003c/p\u003e \u003cp\u003eAccordingly, this study pursues two objectives. First, it conducts a confirmatory validation of the Albanian versions of the GAAIS, AT-EAI, and AIAS, assessing their factorial structure and psychometric reliability. Second, it applies predictive modelling to examine how ethical attitudes and AI-related anxiety explain variation in positive and negative attitudes toward AI, controlling for demographic factors such as gender, age, education level, and exposure to AI-related information. This sequential design\u0026mdash;combining confirmatory factor analysis with hierarchical regression\u0026mdash;aligns with the methodological logic of the original GAAIS validation studies and allows for a robust assessment of both measurement and predictive validity in the Albanian higher-education context (Cicero et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Hubertz and Janowsky, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBy linking ethics, emotion, and attitude formation, this study contributes to the expanding literature on the psychological and moral foundations of technology acceptance (Jang et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Mubashir et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Weng et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). It advances the field in three ways: first, by providing validated Albanian versions of widely used international instruments, enabling cross-cultural research; second, by empirically demonstrating the joint role of ethical cognition and emotional response in shaping positive and negative attitudes toward AI; and third, by offering policy-relevant insights for universities seeking to integrate AI responsibly through curricula that combine technical competence with ethical reflection and emotional preparedness.\u003c/p\u003e \u003cp\u003eThe remainder of the paper is structured as follows. Section 2 reviews the theoretical and empirical literature on ethical and emotional determinants of AI attitudes. Section 3 describes the data, instruments, and analytical strategy. Section 4 presents the empirical results. Section 5 discusses implications for theory, practice, and higher-education policy. Section 6 concludes with limitations and directions for future research.\u003c/p\u003e"},{"header":"2. Literature review","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Measuring Attitudes Toward Artificial Intelligence: The GAAIS Framework\u003c/h2\u003e \u003cp\u003eThe General Attitudes toward Artificial Intelligence Scale (GAAIS), developed by Schepman and Rodway (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), represents one of the most widely adopted instruments for capturing the multidimensional nature of public attitudes toward AI. Rather than treating attitudes as a unidimensional continuum ranging from acceptance to rejection, the GAAIS explicitly distinguishes between positive and negative evaluative dimensions. This conceptualisation reflects growing recognition in the attitude and risk-perception literature that individuals often hold ambivalent or coexisting evaluations of emerging technologies, simultaneously acknowledging their benefits while expressing concern about their societal and personal consequences.\u003c/p\u003e \u003cp\u003eThe positive dimension of the GAAIS captures attitudes related to curiosity, enthusiasm, perceived usefulness, and anticipated societal or personal benefits of AI. Items loading on this factor reflect optimism regarding AI\u0026rsquo;s potential to enhance efficiency, improve well-being, and support human activities. In contrast, the negative dimension captures scepticism, distrust, and perceived risk, encompassing concerns about loss of control, ethical misuse, surveillance, and broader social harms. Importantly, these two dimensions are not conceptualised as simple opposites but as partially independent constructs, allowing individuals to express both high optimism and high concern simultaneously.\u003c/p\u003e \u003cp\u003eThe original validation study conducted in the United Kingdom and Italy reported a robust two-factor structure with strong psychometric properties, including high internal consistency for both the positive and negative subscales (Cicero et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Sacco et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Schepman and Rodway, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Subsequent applications and validations across diverse sociocultural contexts \u0026ndash; including East Asia, Latin America, and the Middle East - have largely replicated this bi-dimensional architecture, supporting the structural stability and cross-cultural relevance of the GAAIS (Lee et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Longoni and Cian, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). These studies consistently show that general attitudes toward AI are best understood as a configuration of concurrent positive expectations and negative concerns, rather than as a single evaluative stance.\u003c/p\u003e \u003cp\u003eAt the same time, cross-national research highlights that the relative salience of positive and negative attitudes varies systematically across cultural, institutional, and technological contexts. In societies characterised by lower levels of institutional trust, limited regulatory oversight, or uneven exposure to advanced digital technologies, negative affective responses \u0026ndash; such as anxiety, uncertainty, and perceived loss of agency \u0026ndash; tend to be more pronounced (Hubertz and Janowsky, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Kaya et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Mantello et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Conversely, in contexts where AI is embedded within stable governance frameworks and transparent institutional environments, positive evaluations related to utility and innovation are more likely to dominate.\u003c/p\u003e \u003cp\u003eThese insights underscore the importance of contextual validation when applying the GAAIS beyond the settings in which it was originally developed. This is particularly relevant for Albania, where attitudes toward automation, digital governance, and algorithmic decision-making are shaped by a combination of rapid digitalisation, ongoing institutional transition, and historically low levels of public trust in formal institutions. In such contexts, perceptions of AI may be more strongly influenced by affective responses, ethical concerns, and informal experiences than by established regulatory assurances. Validating the GAAIS in the Albanian higher-education context is therefore essential to ensure both its psychometric robustness and its conceptual sensitivity to local socio-institutional conditions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Ethical Attitudes Toward Artificial Intelligence: The AT-EAI Scale\u003c/h2\u003e \u003cp\u003e The Attitudes toward Ethical Artificial Intelligence Scale (AT-EAI) developed by Jang et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) operationalises the normative and ethical dimensions of AI acceptance, shifting the focus from instrumental evaluations toward questions of moral legitimacy and societal responsibility. Grounded in international frameworks for trustworthy AI - most notably those articulated by UNESCO and the OECD - the AT-EAI conceptualises ethical attitudes as a set of cognitive evaluations concerning whether AI systems conform to widely endorsed moral principles.\u003c/p\u003e \u003cp\u003eSpecifically, the scale captures five interrelated ethical domains. Fairness refers to the expectation that AI systems should produce non-discriminatory outcomes and avoid bias against individuals or social groups. Transparency concerns the extent to which AI systems are explainable, interpretable, and open to scrutiny by users and affected stakeholders. Non-maleficence reflects the foundational ethical principle of \u0026ldquo;do no harm,\u0026rdquo; encompassing concerns about unintended negative consequences, misuse, or adverse societal impacts. Privacy addresses the protection of personal data and the responsible handling of information in AI-driven processes. Finally, responsibility relates to accountability mechanisms, including clarity over who is answerable for AI-generated decisions and outcomes.\u003c/p\u003e \u003cp\u003eEmpirical validations of the AT-EAI across diverse cultural settings - including Japan, South Korea, and several European countries - have demonstrated strong psychometric properties, with internal consistency coefficients typically ranging between α\u0026thinsp;=\u0026thinsp;0.74 and α\u0026thinsp;=\u0026thinsp;0.84, as well as stable factorial structures across contexts (Kong et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Lim et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Sun and Zhou, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Yilmaz et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zhu et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Beyond measurement reliability, a growing body of research indicates that perceived ethical alignment plays a central role in shaping trust, legitimacy, and acceptance of AI applications (Arora and Garg, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Jobin et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Ethical evaluations thus function as a critical cognitive filter through which individuals assess whether AI technologies are socially acceptable and worthy of integration into sensitive domains.\u003c/p\u003e \u003cp\u003eIn the context of higher education, ethical perceptions are particularly salient. Students and academic staff increasingly interact with algorithmic systems - such as plagiarism detection software, automated grading tools, learning analytics, and adaptive learning platforms - whose decision logic is often opaque and poorly understood (Jang et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Sultana et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). These systems directly affect academic evaluation, surveillance, and data use, amplifying concerns related to fairness, transparency, and accountability. Validating the AT-EAI in the Albanian context therefore extends its applicability to a setting where formal awareness of AI ethics remains limited and regulatory frameworks are still evolving, making ethical cognition a potentially decisive factor in shaping attitudes toward AI in universities.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Anxiety Toward Artificial Intelligence: The AIAS Framework\u003c/h2\u003e \u003cp\u003eWhile ethical attitudes capture the cognitive\u0026ndash;normative evaluation of AI, emotional responses represent an equally important component of attitude formation. The AI Anxiety Scale (AIAS) introduced by (Wang et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Wang and Wang, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) addresses this affective dimension by measuring the extent to which AI technologies evoke feelings of apprehension, unease, or threat. The scale is theoretically grounded in dual-process models of emotion and risk perception, which distinguish between cognitive appraisals of threat and affective emotional reactions (Slovic et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2004\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe AIAS differentiates between two related but distinct forms of anxiety. Cognitive anxiety reflects concerns about understanding AI systems, keeping pace with rapid technological change, and coping with perceived skill obsolescence. It captures feelings of confusion, uncertainty, and perceived inadequacy in the face of increasingly complex AI technologies. Affective anxiety, by contrast, refers to emotional unease and fear, including worries about dependency on AI, loss of human autonomy, or displacement of human roles by intelligent systems (Skagerlund et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Wang and Wang, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).. This two-factor structure has been consistently supported across empirical studies and aligns with established affective\u0026ndash;cognitive models of technological anxiety.\u003c/p\u003e \u003cp\u003ePsychometric evaluations of the AIAS report high internal consistency (typically α\u0026thinsp;\u0026gt;\u0026thinsp;.85) and stable factorial validity across Western and Asian samples (Kaya et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Nomura and Tanaka, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Wang and Wang, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Substantive research using the scale shows that higher levels of AI-related anxiety are associated with lower perceived usefulness, reduced acceptance of AI technologies, and stronger ethical and social concerns. Conversely, exposure to AI education, transparent system design, and opportunities for informed engagement have been shown to attenuate anxiety and foster more favourable attitudes toward AI.\u003c/p\u003e \u003cp\u003eIn higher-education settings, AI anxiety is particularly consequential. Students are often required to engage with AI-enabled systems in evaluative and high-stakes contexts\u0026mdash;such as assessment, academic integrity monitoring, and performance analytics\u0026mdash;where uncertainty and perceived loss of control may heighten emotional responses. Recent evidence suggests that emotional reactions to AI can shape general attitudes as strongly as cognitive evaluations of ethics or utility (Kong et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Xue et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The AIAS is therefore well suited to examining how affective anxiety interacts with ethical cognition to shape positive and negative attitudes toward AI in higher education, particularly in transitional contexts such as Albania, where rapid digitalisation coexists with limited institutional reassurance.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4. The Albanian Context: Digitalisation, Youth, and Higher Education\u003c/h2\u003e \u003cp\u003eDigitalisation has emerged as a central pillar of Albania\u0026rsquo;s development agenda, framed simultaneously as an educational reform imperative and a strategic response to youth labour-market challenges. In the education sector, the National Strategy for Education 2021\u0026ndash;2026 identifies digital competence as a core transversal objective (Objective A4), emphasising the systematic integration of information and communication technologies (ICT) across teaching, learning, and institutional management (GoA, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; MASR, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This policy orientation reflects broader efforts to align Albania\u0026rsquo;s education system with European standards and to equip students with skills relevant to a rapidly transforming digital economy.\u003c/p\u003e \u003cp\u003eDespite these ambitions, structural constraints continue to shape the lived realities of digitalisation in education. Persistent infrastructural gaps remain evident, particularly in earlier stages of the education system. As reported by the Council of Ministers (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) and Ministry of Education and Sports (MAS, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), the computer-to-student ratio in pre-university education stood at approximately 1:27 in 2017, a figure markedly below the European Union benchmark range of 1:3 to 1:7. While conditions improve at the tertiary level, these disparities signal cumulative inequalities in digital exposure that accompany students into higher education. To mitigate such gaps, the Albanian Academic Network (RASH) plays a key enabling role by providing universities with access to learning management systems, digital repositories, research databases, and blended-learning infrastructures, thereby supporting the gradual digital modernisation of higher-education institutions (Council of Ministers, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Recent national policy documents explicitly frame AI adoption around trustworthy and human-centred use, highlighting ethical safeguards, transparency, data governance, and alignment with emerging EU regulatory approaches. This macro-policy emphasis makes universities a critical implementation arena, where student acceptance is likely to hinge on whether AI systems are experienced as safe (non-maleficence) and rights-protective (privacy and accountability), rather than merely efficient. Against this background, examining ethical endorsement and anxiety as joint predictors of AI attitudes offers an empirically grounded indicator of Albania\u0026rsquo;s \u0026lsquo;readiness\u0026rsquo; for responsible AI integration in higher education (GoA, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eParallel developments are evident in the labour-market and youth-policy domains. The Youth Guarantee (YG) initiative, implemented in Albania with EU support, explicitly integrates digital tools to engage young people not in employment, education, or training (NEETs). The YG\u0026rsquo;s online platform - embedded within the national e-Albania portal - and the forthcoming dedicated Digital Portal aim to streamline registration, profiling, counselling, and referral to training or employment opportunities (EU4Youth, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; European Commission, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2025a\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2025b\u003c/span\u003e). These efforts are reinforced by the National Youth Strategy 2022\u0026ndash;2029, which prioritises digital literacy, online career guidance, and technology-mediated access to services as mechanisms for improving youth employability and social inclusion. In parallel, the expansion of digital freelancing and remote work has enabled segments of Albanian youth to participate in global labour markets, often bypassing domestic institutional constraints (ETF, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHowever, the rapid diffusion of digital tools has not been matched by equivalent progress in institutional capacity, pedagogical preparedness, or ethical governance, particularly within higher education. Universities continue to face challenges related to uneven digital infrastructure, limited staff training in advanced digital and AI-enabled tools, and the absence of systematic frameworks for ethical reflection on algorithmic systems. As a result, students\u0026rsquo; encounters with AI technologies - ranging from learning platforms and plagiarism detection software to generative AI applications - are frequently mediated through informal learning, peer experimentation, and fragmented institutional guidance.\u003c/p\u003e \u003cp\u003eIn this context, attitudes toward AI among Albanian students are likely to be shaped by a combination of unequal digital access, experiential uncertainty, and limited exposure to structured ethical discourse. Ethical awareness and AI-related anxiety thus become especially salient lenses through which to examine how digital transformation is cognitively and emotionally internalised by young people preparing to enter both national and transnational labour markets. Analysing these dimensions within the Albanian higher-education context offers valuable insights into the socio-psychological foundations of AI acceptance in emerging economies, where technological acceleration often outpaces institutional regulation and ethical governance.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5. Theoretical and Empirical Foundations for Predictive Modelling\u003c/h2\u003e \u003cp\u003eTo account for the multidimensional structure of attitudes toward artificial intelligence, the present study adopts an integrated cognitive\u0026ndash;affective\u0026ndash;ethical framework that synthesises three complementary theoretical traditions: (i) affective\u0026ndash;cognitive models of attitude formation, (ii) the affect heuristic and dual-process theories of risk perception, and (iii) extended technology acceptance models (TAM and UTAUT2). Together, these perspectives provide a coherent theoretical rationale for examining how ethical cognition and emotional responses jointly shape general attitudes toward AI and justify their inclusion as key predictors in the empirical model.\u003c/p\u003e \u003cp\u003eFirst, affective\u0026ndash;cognitive models of attitude formation conceptualise attitudes as composite evaluations emerging from the interaction between cognitive beliefs and affective reactions (Ajzen and Fishbein, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Eagly and Chaiken, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e1993\u003c/span\u003e; Liska, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e1984\u003c/span\u003e). Attitudes are not reducible to either rational judgment or emotional response alone; rather, they reflect the dynamic integration of both components. In the present study, ethical cognition, operationalised through the Attitudes toward Ethical Artificial Intelligence Scale (AT-EAI), represents the cognitive evaluative dimension. It captures individuals\u0026rsquo; reasoned assessments of AI systems in terms of fairness, transparency, non-maleficence, privacy, and responsibility. AI anxiety, measured by the AI Anxiety Scale (AIAS), represents the affective dimension of attitude formation, encompassing both cognitive anxiety (concerns about understanding and keeping pace with AI developments) and affective anxiety (emotional unease related to dependence, loss of control, or human replacement) (Wang et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Wang and Wang, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The interaction between these cognitive and affective components constitutes the attitudinal core reflected in the General Attitudes toward Artificial Intelligence Scale (GAAIS).\u003c/p\u003e \u003cp\u003eSecond, insights from the affect heuristic and dual-process theories of risk perception further clarify why ethical evaluations and emotional responses should be modelled jointly rather than independently. Dual-process frameworks posit that cognition and affect operate as parallel systems that jointly inform judgment, particularly under conditions of uncertainty and complexity (Slovic et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). In such contexts, affective reactions often serve as heuristic shortcuts that shape perceptions of risk, trust, and acceptability more rapidly and sometimes more strongly than deliberative reasoning. AI technologies, characterised by opacity, autonomy, and moral ambiguity, are especially prone to affect-driven evaluation. Empirical research shows that emotions such as fear, unease, or moral concern can amplify perceived risk and suppress acceptance even when perceived benefits are acknowledged (Schepman and Rodway, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Sultana et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Consequently, ethical reassurance (as a cognitive signal of legitimacy) and AI-related anxiety (as an affective signal of threat or uncertainty) are expected to interact in shaping overall judgments of AI\u0026rsquo;s desirability and trustworthiness.\u003c/p\u003e \u003cp\u003eThird, this study builds on and extends insights from the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT/UTAUT2), which traditionally explain technology adoption through perceived usefulness, ease of use, and facilitating conditions (Sun and Zhou, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Venkatesh et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). While these frameworks have proven robust for explaining adoption of conventional information systems, recent empirical work highlights their limitations when applied to AI technologies, which raise broader ethical, social, and emotional concerns. Extensions of TAM and UTAUT increasingly incorporate trust, moral evaluation, and emotional comfort as central determinants of acceptance, particularly in high-stakes or socially sensitive domains (Longoni and Cian, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Nomura and Tanaka, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). From this perspective, general attitudes toward AI are better understood as the outcome of an evaluative process that integrates ethical judgment and emotional response, rather than as a direct function of instrumental utility alone.\u003c/p\u003e \u003cp\u003eEmpirically, this integrated framework leads to clear expectations for predictive modelling. Individuals who perceive AI systems as ethically legitimate, fair, and accountable are expected to report stronger positive attitudes and weaker negative attitudes toward AI. Conversely, higher levels of AI-related anxiety, whether cognitive or affective, are expected to dampen optimism and amplify scepticism, risk sensitivity, and negative evaluations. Ethical cognition and emotional response are therefore conceptualised as dual antecedents of general attitudes toward AI, exerting independent and potentially opposing effects on positive and negative attitude dimensions.\u003c/p\u003e \u003cp\u003eThe resulting conceptual pathway underpinning the empirical analysis can be summarised as follows:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:Ethical\\:Cognition\\:\\left(Cognitive\\:Appraisal\\right)\\:+\\:AI\\:Anxiety\\:\\left(Affective\\:Reaction\\right)\\:\\to\\:\\:General\\:Attitudes\\:toward\\:AI\\:\\left(Positive\\:and\\:Negative\\:Evaluations\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eEq.\u0026nbsp;1\u003c/p\u003e \u003cp\u003eThis theoretical synthesis provides the foundation for the study\u0026rsquo;s predictive modelling strategy, which empirically tests these relationships using hierarchical regression analyses following the confirmatory validation of the GAAIS, AT-EAI, and AIAS instruments. By embedding predictive modelling within a clearly articulated socio-psychological framework, the study moves beyond descriptive validation and offers a theory-driven explanation of how ethical and emotional mechanisms jointly shape attitudes toward AI in higher education.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Research Hypotheses","content":"\u003cp\u003eBuilding on affective-cognitive models of attitude formation, dual-process theories of risk perception, and extended technology acceptance frameworks, the study formulates the following hypotheses concerning the ethical and emotional determinants of students\u0026rsquo; attitudes toward artificial intelligence in higher education.\u003c/p\u003e \u003cp\u003eH1a (Ethical cognition and positive attitudes).\u003c/p\u003e \u003cp\u003e Stronger endorsement of ethical principles guiding AI (AT-EAI dimensions) is positively associated with Positive GAAIS scores, reflecting greater optimism and perceived benefits of AI.\u003c/p\u003e \u003cp\u003eH1b (Ethical cognition and negative attitudes).\u003c/p\u003e \u003cp\u003e Stronger endorsement of ethical principles guiding AI is negatively associated with Negative GAAIS scores, indicating reduced scepticism and perceived risk; however, privacy concerns may be positively associated with Negative GAAIS by heightening sensitivity to data-related risks.\u003c/p\u003e \u003cp\u003eH2a (Cognitive anxiety and positive attitudes).\u003c/p\u003e \u003cp\u003eHigher levels of cognitive AI anxiety are negatively associated with Positive GAAIS scores.\u003c/p\u003e \u003cp\u003eH2b (Affective anxiety and positive attitudes).\u003c/p\u003e \u003cp\u003eHigher levels of affective AI anxiety are negatively associated with Positive GAAIS scores.\u003c/p\u003e \u003cp\u003eH3a (Cognitive anxiety and negative attitudes).\u003c/p\u003e \u003cp\u003eHigher levels of cognitive AI anxiety are positively associated with Negative GAAIS scores.\u003c/p\u003e \u003cp\u003eH3b (Affective anxiety and negative attitudes).\u003c/p\u003e \u003cp\u003eHigher levels of affective AI anxiety are positively associated with Negative GAAIS scores.\u003c/p\u003e \u003cp\u003eTogether, these hypotheses reflect an integrated cognitive\u0026ndash;affective\u0026ndash;ethical model in which ethical evaluations and emotional responses function as distinct but complementary antecedents of both optimistic and risk-oriented attitudes toward artificial intelligence.\u003c/p\u003e"},{"header":"4. Methods","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Participants and procedure\u003c/h2\u003e \u003cp\u003eThis study employed a cross-sectional survey design conducted between February and April 2025 at the University of Shkodra \u0026ldquo;Luigj Gurakuqi,\u0026rdquo; Albania. Data were collected using a structured questionnaire administered both in-class and via an online Qualtrics link to ensure broad participation across study programmes. After screening for incomplete questionnaires and patterned or inattentive responses, a total of 705 valid observations were retained for analysis.\u003c/p\u003e \u003cp\u003eParticipants were drawn from a wide range of academic programmes, including business, education, information sciences, and social sciences, broadly reflecting the university\u0026rsquo;s enrolment structure. The sample was 71.4% female and 28.6% male, consistent with the gender composition of Albanian higher education, particularly in the social sciences and education fields (MASR, 2024). With respect to age, 56.9% of respondents were aged 18\u0026ndash;25, 25.5% were 26\u0026ndash;35, and 17.6% were 36\u0026ndash;50, indicating the inclusion of both traditional students and mature learners. Approximately half of the respondents were enrolled at the undergraduate level, while master\u0026rsquo;s and other postgraduate students constituted the remainder.\u003c/p\u003e \u003cp\u003eThe Albanian higher-education context is particularly appropriate for examining attitudes toward artificial intelligence. Students and academic staff increasingly encounter AI-enabled tools-such as generative AI applications, plagiarism detection systems, and learning-management-system analytics-within an institutional environment where frameworks for data protection, algorithmic transparency, and AI ethics are still evolving. Participation was voluntary, informed consent was obtained prior to survey completion, and no incentives were offered.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.2. Measures\u003c/h2\u003e \u003cp\u003eAll constructs were measured using internationally validated instruments, translated and adapted into Albanian following standard translation\u0026ndash;back-translation procedures to ensure conceptual equivalence. Unless otherwise specified, items were rated on five-point Likert scales (1\u0026thinsp;=\u0026thinsp;strongly disagree, 5\u0026thinsp;=\u0026thinsp;strongly agree).\u003c/p\u003e \u003cp\u003eEthical evaluations of AI were measured using the 17-item AT-EAI developed by Jang et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The scale captures five core dimensions of trustworthy AI: fairness, transparency, non-maleficence, privacy, and responsibility. Example items include \u0026ldquo;AI should be designed to avoid bias or discrimination\u0026rdquo; (fairness) and \u0026ldquo;People should be informed when AI influences outcomes\u0026rdquo; (transparency). Reverse-coded items were recoded so that higher scores reflected stronger endorsement of ethical AI principles. In the present sample, internal consistency was satisfactory across all subscales (α\u0026thinsp;=\u0026thinsp;0.74\u0026ndash;0.83), consistent with prior validation studies.\u003c/p\u003e \u003cp\u003eEmotional responses to AI were assessed using the AIAS (Wang et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Wang and Wang, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), which consists of 14 items loading on two factors: cognitive anxiety (difficulty understanding or keeping up with AI developments) and affective anxiety (fear of dependence, loss of control, or role replacement). To preserve comparability with the original instrument and retain response variability, the original seven-point Likert scale (1\u0026thinsp;=\u0026thinsp;strongly disagree, 7\u0026thinsp;=\u0026thinsp;strongly agree) was maintained. Example items include \u0026ldquo;Learning to use AI products makes me anxious\u0026rdquo; (cognitive anxiety) and \u0026ldquo;I fear AI could replace human workers\u0026rdquo; (affective anxiety). Reliability was excellent (overall α\u0026thinsp;=\u0026thinsp;0.91; cognitive α\u0026thinsp;=\u0026thinsp;0.88; affective α\u0026thinsp;=\u0026thinsp;0.86).\u003c/p\u003e \u003cp\u003eGeneral attitudes toward AI were measured using GAAIS scale, developed by Schepman and Rodway (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The Albanian version retained the original two-factor structure: (i) Positive GAAIS (12 items), capturing optimism, interest, and perceived societal benefits (e.g., \u0026ldquo;Much of society will benefit from a future with AI\u0026rdquo;); and (ii) Negative GAAIS (8 items), capturing discomfort, concern, and perceived danger (e.g., \u0026ldquo;AI is dangerous\u0026rdquo;). Both subscales demonstrated high internal consistency in the present sample (α\u0026thinsp;=\u0026thinsp;0.89 for Positive; α\u0026thinsp;=\u0026thinsp;0.83 for Negative), in line with confirmatory factor analysis results reported later.\u003c/p\u003e \u003cp\u003eTo isolate the predictive effects of ethical attitudes and AI anxiety, several background variables were included as controls: gender (1\u0026thinsp;=\u0026thinsp;male), age group (18\u0026ndash;25; 26\u0026ndash;35; 36\u0026ndash;50 as reference), educational status (bachelor student, bachelor graduate, master student, master graduate; PhD/other as reference), and self-reported knowledge of AI (no information, little, sufficient, extensive). Birth order (only child, first-born, middle, youngest) was also recorded to explore potential family-structure effects, but was treated as a secondary control.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e4.3. Ethical Considerations\u003c/h2\u003e \u003cp\u003e The study was conducted in accordance with the ethical standards of the University of Shkodra \u0026ldquo;Luigj Gurakuqi.\u0026rdquo; Ethical approval was granted by the University Ethics Committee (Approval No. 2025/AEI-05). Participants were informed about the study\u0026rsquo;s objectives, assured of anonymity and confidentiality, and informed of their right to withdraw at any time without consequence. No personally identifying information was collected.\u003c/p\u003e \u003cp\u003eAll instruments underwent translation and back-translation procedures, and a pilot study with 35 students was conducted to assess clarity, cultural appropriateness, and completion time. Minor linguistic refinements\u0026mdash;primarily simplifying items related to transparency and responsibility\u0026mdash;were implemented prior to full deployment. Data were stored on password-protected institutional servers and used exclusively for research purposes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e4.4. Descriptive Analysis of the Sample\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents descriptive statistics for the 705 respondents included in the analysis. The demographic composition closely mirrors both the structure of the University of Shkodra \u0026ldquo;Luigj Gurakuqi\u0026rdquo; and broader participation patterns in Albanian higher education. A pronounced gender imbalance is evident, with women constituting 71.4% of the sample. This pattern aligns with national trends indicating female predominance in education, business, and social-science programmes (GoA, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; MAS, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; MASR, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), reflecting the broader feminisation of tertiary education in Albania. In terms of age, the majority of respondents (56.9%) were aged 18\u0026ndash;25, representing traditional undergraduate students. A further 25.5% were aged 26\u0026ndash;35, typically postgraduate students, while 17.6% were aged 36\u0026ndash;50, capturing mature learners and doctoral candidates. This distribution enables generational comparisons in exposure to digital technologies and familiarity with AI. Educational status was heterogeneous: 30% were bachelor\u0026rsquo;s students, 14.7% bachelor graduates, 18% master\u0026rsquo;s students, and 34% master\u0026rsquo;s graduates, with 3.1% reporting doctoral-level engagement. This mix ensures representation across academic stages, from early users of digital learning systems to advanced researchers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptive Statistics of Sample Characteristics (N\u0026thinsp;=\u0026thinsp;705)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eObs.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMin\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMax\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.286\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.452\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAge group\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e18\u0026ndash;25 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.568\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.495\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e26\u0026ndash;35 years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.255\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.436\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e36\u0026ndash;50 years (ref. category)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.175\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.381\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eEducational status\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBachelor student\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.300\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.458\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBachelor graduate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.147\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.354\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaster student\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.180\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.384\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaster graduate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.340\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.474\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePhD holder (ref. category)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.174\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eSelf-reported knowledge of AI\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo information (ref. category)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.090\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.287\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLittle information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.546\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.498\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSufficient information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.473\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExtensive information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.157\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003eNote. All variables are dummy-coded (0\u0026thinsp;=\u0026thinsp;No, 1\u0026thinsp;=\u0026thinsp;Yes). M\u0026thinsp;=\u0026thinsp;mean; SD\u0026thinsp;=\u0026thinsp;standard deviation. All statistics are based on N\u0026thinsp;=\u0026thinsp;705 valid observations.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eRegarding self-reported AI knowledge, 54.6% of respondents reported little familiarity with AI, 33.7% sufficient knowledge, 2.5% extensive knowledge, and 9.0% indicated no prior information. This pattern suggests broad awareness but uneven depth of understanding \u0026ndash; an important condition for analysing how ethical evaluations and anxiety jointly shape attitudes toward AI. Comparisons with (INSTAT, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) data (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) indicate that the sample broadly reflects Albania\u0026rsquo;s higher-education population in terms of gender, age, and level of study, albeit with a modest overrepresentation of postgraduate students.\u003c/p\u003e \u003cp\u003eThis comparability supports the external validity of the findings and suggests that observed attitudinal patterns are likely indicative of broader tendencies among Albanian university students. In comparative terms, the study sample aligns closely with national higher-education demographics reported by INSTAT (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Although women are slightly overrepresented in the sample (71.4% vs. 59% nationally), this reflects the disciplinary profile of the University of Shkodra, where enrolment is concentrated in education, business, and social sciences\u0026mdash;fields characterised by higher female participation.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparative profile of higher-education students in Albania, University of Shkodra (UNISHK), and study sample (2024)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCharacteristic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNational higher-education population (%)\u0026sup1;\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eUNISHK student population (%)\u0026sup2;\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eStudy sample (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eComment\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFemale 59 Male 41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFemale\u0026thinsp;\u0026asymp;\u0026thinsp;68 Male\u0026thinsp;\u0026asymp;\u0026thinsp;32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFemale 71.4 Male 28.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSlight female over-representation, consistent with social-science and education dominance\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLevel of study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBachelor 55 Master 38 Doctorate 1 Short cycle 6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBachelor\u0026thinsp;\u0026asymp;\u0026thinsp;60 Master\u0026thinsp;\u0026asymp;\u0026thinsp;35 Doctorate\u0026thinsp;\u0026asymp;\u0026thinsp;2 Short cycle\u0026thinsp;\u0026asymp;\u0026thinsp;3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBachelor 45 Master 52 Doctorate 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHigher share of postgraduates, reflecting graduate research focus\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eType of institution\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePublic 71 Private 29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100 Public\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e100 Public\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSample drawn entirely from a public university\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge distribution\u0026sup3;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMajority 18\u0026ndash;25 (\u0026asymp;\u0026thinsp;65\u0026ndash;70%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18\u0026ndash;25\u0026thinsp;\u0026asymp;\u0026thinsp;60 26\u0026ndash;35\u0026thinsp;\u0026asymp;\u0026thinsp;25 36\u0026thinsp;+\u0026thinsp;\u0026asymp;\u0026thinsp;15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18\u0026ndash;25 56.9 26\u0026ndash;35 25.5 36\u0026ndash;50 17.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eComparable youth concentration with inclusion of mature learners\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal enrolment (N)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e122 613 students\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026asymp;\u0026thinsp;6 000 students\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e705 respondents\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e\u0026sup1; Source: Institute of Statistics (INSTAT), Student\u0026euml; t\u0026euml; regjistruar n\u0026euml; arsimin e lart\u0026euml; sipas programeve t\u0026euml; studimit, 2024.\u003c/p\u003e \u003cp\u003e\u0026sup2; Approximate distribution based on internal University of Shkodra \u0026ldquo;Luigj Gurakuqi\u0026rdquo; enrolment statistics, 2024.\u003c/p\u003e \u003cp\u003e\u0026sup3; Age breakdown estimated from enrolment by cycle and typical entry age.\u003c/p\u003e \u003cp\u003eSimilarly, while bachelor students represent the majority nationally (55%), the sample includes a somewhat higher proportion of master\u0026rsquo;s students (52%), consistent with the university\u0026rsquo;s research-oriented focus and strong graduate enrolment. Age composition also mirrors national patterns, with enrolment concentrated in the 18\u0026ndash;25 age group (56.9%), alongside meaningful representation of 26\u0026ndash;35 (25.5%) and 36\u0026ndash;50 (17.6%) cohorts. This generational spread enables analysis across different stages of academic and professional experience. Overall, the demographic structure of the 705 respondents closely reflects the broader Albanian higher-education population\u0026mdash;female-dominated, youth-centred, and primarily bachelor- and master-level \u0026ndash; while retaining sufficient postgraduate and mature-learner representation. This comparability supports the external validity of the findings and suggests that the observed ethical and emotional patterns toward AI are likely indicative of broader trends among university students in Albania.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Confirmatory factor analysis","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e5.1. Data analysis\u003c/h2\u003e \u003cp\u003eAll analyses were conducted in Stata 18. Item-level descriptive statistics (mean, standard deviation, skewness, kurtosis) were inspected to assess distributional properties. Values of skewness and kurtosis within the range \u0026minus;\u0026thinsp;2 to +\u0026thinsp;2 were treated as acceptable for univariate normality (Kline, 2016), supporting the use of parametric procedures.\u003c/p\u003e \u003cp\u003eTo evaluate factorial validity, confirmatory factor analyses (CFA) were estimated in Stata\u0026rsquo;s SEM framework using maximum likelihood with robust standard errors (MLR). For each instrument \u0026ndash; GAAIS, AT-EAI, and AIAS \u0026ndash; two competing specifications were tested: (i) a one-factor model and (ii) the theoretically specified multi-factor model (two-factor for GAAIS, five-factor for AT-EAI, two-factor for AIAS). To ensure comparability across predictors, all scale scores used in regression were computed as mean indices of their respective items; AT-EAI and GAAIS remain on 1\u0026ndash;5 metrics, while AIAS remains on a 1\u0026ndash;7 metric, and coefficients are therefore interpreted as partial associations within each scale\u0026rsquo;s natural unit (standardised βs are reported for cross-predictor comparison).\u003c/p\u003e \u003cp\u003eModel fit was assessed using χ\u0026sup2;/df (acceptable range 1\u0026ndash;3), CFI and TLI (\u0026ge;\u0026thinsp;0.90, preferably\u0026thinsp;\u0026ge;\u0026thinsp;0.95), and RMSEA and SRMR (\u0026le;\u0026thinsp;0.08 indicating good fit). Model parsimony was compared using the AIC, where lower values indicate better fit and differences\u0026thinsp;\u0026gt;\u0026thinsp;2 suggest meaningful improvement.\u003c/p\u003e \u003cp\u003eInternal reliability was assessed using Cronbach\u0026rsquo;s α and McDonald\u0026rsquo;s ω, with \u0026ge;\u0026thinsp;0.70 indicating acceptable consistency. Composite reliability (CR) and average variance extracted (AVE) were also computed to evaluate convergent validity (CR\u0026thinsp;\u0026gt;\u0026thinsp;0.70; AVE\u0026thinsp;\u0026gt;\u0026thinsp;0.50). Across the three scales, reliability and validity metrics met or exceeded recommended thresholds, supporting the adequacy of the Albanian versions for subsequent predictive modelling.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e5.2. Findings\u003c/h2\u003e \u003cp\u003eThe first study aimed to verify whether the three instruments\u0026mdash;the General Attitudes toward Artificial Intelligence Scale (GAAIS), the Attitudes toward Ethical Artificial Intelligence Scale (AT-EAI), and the AI Anxiety Scale (AIAS)\u0026mdash;retain their original factorial structure when applied to Albanian higher-education students. Overall, the CFAs confirmed very good model fit for all three measures, with fit indices meeting or exceeding conventional thresholds, and with factor loadings that were statistically significant and of satisfactory magnitude.\u003c/p\u003e \u003cp\u003e \u003cem\u003eGeneral Attitudes toward AI (GAAIS)\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe hypothesised two-factor structure (Positive vs. Negative attitudes) fit the data very well: χ\u0026sup2; (169)\u0026thinsp;=\u0026thinsp;228.41, p\u0026thinsp;=\u0026thinsp;0.003; χ\u0026sup2;/df\u0026thinsp;=\u0026thinsp;1.35; CFI\u0026thinsp;=\u0026thinsp;0.987; TLI\u0026thinsp;=\u0026thinsp;0.986; RMSEA\u0026thinsp;=\u0026thinsp;0.031 (90% CI [0.018, 0.044]); SRMR\u0026thinsp;=\u0026thinsp;0.064. All items loaded significantly (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) on their intended factors, confirming that a unidimensional solution is unnecessary. For Positive GAAIS, standardized loadings were moderate-to-high (0.48\u0026ndash;0.80). The strongest indicators emphasised enthusiasm and anticipated societal benefit, including \u0026ldquo;AI is exciting\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.798), \u0026ldquo;Much of society will benefit from a future with AI\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.769), and \u0026ldquo;I am interested in using AI systems in my daily life\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.739).\u003c/p\u003e \u003cp\u003eItems reflecting comparative performance claims (e.g., outperforming humans in routine jobs) loaded somewhat lower (\u0026asymp;\u0026thinsp;0.52\u0026ndash;0.53), suggesting that positivity in this sample is driven more by future-oriented optimism and engagement than by narrowly productivity-based beliefs. Internal consistency was high (α\u0026thinsp;=\u0026thinsp;0.887). Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e presents the standardized factor loadings, estimates, and item-level descriptive statistics for the Positive and Negative GAAIS dimensions.\u003c/p\u003e \u003cp\u003e\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt; Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e here \u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eConfirmatory Factor Loadings for the Albanian GAAIS\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eItem (indicator)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEstimate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ez\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e95% CI Lower\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e95% CI Upper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eStandardized Estimate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eSD\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePositive GAAIS\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFor routine transactions, I would rather interact with an AI\u003c/p\u003e \u003cp\u003e system than a human.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.602\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e19.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.540\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.663\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.482\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.512\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.142\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI can provide new economic opportunities for this country.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.656\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.600\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.713\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.604\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.984\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.962\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI systems can help people feel happier.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.529\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.480\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.578\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.548\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.226\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.876\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI am impressed by what AI can do.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.473\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.026\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e17.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.420\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.525\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.520\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.286\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.812\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI am interested in using AI systems in my daily life.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.841\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e24.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.773\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.910\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.739\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.567\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.053\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI can have positive impacts on people\u0026rsquo;s well-being.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.648\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.027\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.594\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.702\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.714\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.874\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.861\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI is exciting.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.711\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.653\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.770\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.798\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.182\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.904\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAn AI agent would outperform humans in many routine jobs.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.597\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e18.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.534\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.661\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.526\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.224\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.156\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThere are many beneficial applications of AI.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.558\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.509\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.607\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.682\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.351\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.744\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI systems can perform better than humans.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.535\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e18.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.479\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.591\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.522\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.476\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.957\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMuch of society will benefit from a future with AI.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.864\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.035\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e24.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.795\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.934\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.769\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.583\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.042\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI would like to use AI in my own job.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.801\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.728\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.874\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.661\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.978\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.231\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eNegative GAAIS\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOrganizations use AI unethically.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.428\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0316\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.367\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.490\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.413\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.764\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.984\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI systems make many errors.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.312\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.256\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.367\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.308\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.876\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.942\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI find AI sinister.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.942\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.041\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.872\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.861\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.854\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.426\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.129\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI might take control of people.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.731\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.038\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e19.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.656\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.807\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.603\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.314\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.188\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI is dangerous.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.817\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.039\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.740\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.894\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.694\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.012\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.166\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI shiver with discomfort when thinking about AI\u0026rsquo;s future uses.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.038\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.042\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e24.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.954\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.123\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.826\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.496\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.247\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePeople like me will suffer if AI is used more and more.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.709\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e18.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.635\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.783\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.566\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.152\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.175\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI is used to spy on people.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.577\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.035\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e16.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.507\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.646\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.472\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.754\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.132\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFor Negative GAAIS, several items showed very high loadings, particularly those capturing visceral discomfort and perceived threat, such as \u0026ldquo;I find AI sinister\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.854) and \u0026ldquo;I shiver with discomfort when thinking about AI\u0026rsquo;s future uses\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.826). Items reflecting broader institutional scepticism (e.g., \u0026ldquo;Organizations use AI unethically\u0026rdquo;, λ\u0026thinsp;=\u0026thinsp;0.413; \u0026ldquo;AI is used to spy on people\u0026rdquo;, λ\u0026thinsp;=\u0026thinsp;0.472) also loaded significantly but were weaker, indicating that affective threat perceptions dominate the negative dimension more than purely institutional distrust.\u003c/p\u003e \u003cp\u003eReliability was good (α\u0026thinsp;=\u0026thinsp;0.829). The two factors were moderately correlated (r\u0026thinsp;=\u0026thinsp;0.392, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), consistent with ambivalent technology attitudes: respondents may simultaneously recognise AI\u0026rsquo;s potential while remaining concerned about risks and loss of control.\u003c/p\u003e \u003cp\u003e \u003cem\u003eAttitudes toward Ethical AI (AT-EAI)\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe five-factor model of the AT-EAI also demonstrated an excellent fit to the data: χ\u0026sup2;(413)\u0026thinsp;=\u0026thinsp;502.16, p\u0026thinsp;=\u0026thinsp;0.072; χ\u0026sup2;/df\u0026thinsp;=\u0026thinsp;1.22; CFI\u0026thinsp;=\u0026thinsp;0.989; TLI\u0026thinsp;=\u0026thinsp;0.987; RMSEA\u0026thinsp;=\u0026thinsp;0.024 (90% CI [0.011, 0.037]); SRMR\u0026thinsp;=\u0026thinsp;0.041. These results confirm that Albanian students are able to differentiate between distinct ethical dimensions of AI-fairness, transparency, non-maleficence, privacy, and responsibility-rather than treating ethics as a single undifferentiated concern.\u003c/p\u003e \u003cp\u003eAll items loaded significantly (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) on their intended dimensions and most standardized loadings were comfortably above 0.70. The transparency and responsibility factors showed some of the strongest loadings (e.g. \u0026ldquo;AI developers should disclose how algorithms make decisions\u0026rdquo;, λ\u0026thinsp;=\u0026thinsp;0.802; \u0026ldquo;There should be clear laws assigning responsibility for AI errors\u0026rdquo;, λ\u0026thinsp;=\u0026thinsp;0.816), which suggests that, in this context, students place particular weight on explainability and accountability in AI systems. This is coherent with the institutional situation in Albania, where rules on AI use, data sharing, and algorithmic oversight are still being shaped, so students appear to demand clearer lines of responsibility. Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e reports the standardized factor loadings and item-level descriptive statistics for each of the five AT-EAI subscales.\u003c/p\u003e \u003cp\u003e\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt; Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e here \u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eConfirmatory Factor Loadings for the Albanian AT-EAI Scale\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eItem (indicator)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEstimate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ez\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e95% CI Lower\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e95% CI Upper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eStandardized Estimate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eSD\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eFairness\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI systems should make decisions that treat all people equally.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.641\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20.86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.580\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.702\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.613\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.78\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI should be designed to avoid bias or discrimination.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.712\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e25.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.657\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.767\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.744\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI applications must ensure fairness across social groups.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.681\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.624\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.738\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.711\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eTransparency\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThe decisions made by AI should be explainable to users.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.728\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.030\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e24.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.669\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.787\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.751\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.81\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI developers should disclose how algorithms make decisions.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.816\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e25.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.753\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.879\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.802\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.85\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePeople should be informed when AI influences outcomes.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.688\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.631\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.745\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.703\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.88\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTransparency improves trust in AI systems.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.762\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e24.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.701\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.823\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.771\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eNon-maleficence\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI should never harm or endanger humans.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.812\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.035\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.744\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.880\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.795\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.72\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSafety checks must prevent harmful AI behavior.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.755\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.033\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.690\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.820\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.768\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.75\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI must protect users from unintended negative effects.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.643\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.030\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.584\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.702\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.675\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.78\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAvoiding harm should be the primary goal of AI regulation.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.701\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.640\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.762\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.712\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.77\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePrivacy\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI systems must protect users\u0026rsquo; personal data.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.721\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.658\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.735\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eData collected by AI should not be shared without consent.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.803\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.737\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.869\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.781\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUsers should be able to control what data AI collects.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.677\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.616\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.738\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.695\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eResponsibility\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDevelopers should be held accountable for AI outcomes.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.724\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.030\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e24.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.666\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.782\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.744\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThere should be clear laws assigning responsibility for AI errors.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.853\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.783\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.923\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.816\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.77\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUsers of AI systems must act responsibly.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.672\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.77\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.617\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.727\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.701\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e4.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.83\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe non-maleficence factor was also very robust (λ\u0026thinsp;=\u0026thinsp;0.643\u0026ndash;0.812; α\u0026thinsp;=\u0026thinsp;0.83), indicating that \u0026ldquo;do no harm\u0026rdquo; remains a salient normative anchor even among young users who otherwise welcome AI. Privacy items loaded between 0.677 and 0.803, and fairness items between 0.641 and 0.712, both in acceptable-to-strong ranges. Cronbach\u0026rsquo;s α coefficients ranged from 0.74 (Fairness) to 0.83 (Non-maleficence), all above the recommended 0.70 threshold, confirming internal reliability. Inter-factor correlations were positive and moderate (e.g. r_{Fairness\u0026ndash;Transparency} = 0.41; r_{Fairness\u0026ndash;Responsibility} = 0.31; all p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), suggesting that students who endorse one ethical aspect of AI tend to endorse the others as well, but that these are still empirically separable constructs. This is important for Study 2, because it justifies entering the five AT-EAI subscales separately into the hierarchical regression to see which specific ethical dimension actually predicts AI attitudes.\u003c/p\u003e \u003cp\u003e \u003cem\u003eAI Anxiety Scale (AIAS)\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe confirmatory factor analysis supported the theorised two-factor structure of the AI Anxiety Scale (AIAS), distinguishing cognitive AI anxiety from affective AI anxiety, with excellent global model fit: χ\u0026sup2;(76)\u0026thinsp;=\u0026thinsp;112.84, p\u0026thinsp;=\u0026thinsp;0.004; χ\u0026sup2;/df\u0026thinsp;=\u0026thinsp;1.48; CFI\u0026thinsp;=\u0026thinsp;0.993; TLI\u0026thinsp;=\u0026thinsp;0.991; RMSEA\u0026thinsp;=\u0026thinsp;0.026 (90% CI [0.015, 0.036]); SRMR\u0026thinsp;=\u0026thinsp;0.029. All items loaded significantly (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) on their intended latent factors.\u003c/p\u003e \u003cp\u003eCognitive-anxiety items exhibited strong standardized loadings (λ\u0026thinsp;=\u0026thinsp;0.705\u0026ndash;0.784), with the strongest indicators reflecting difficulty keeping pace with rapid AI developments, such as \u0026ldquo;I feel stressed when I cannot follow AI advances\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.784) and \u0026ldquo;Being unable to keep up with AI makes me anxious\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.758). This pattern indicates that, for Albanian students, cognitive anxiety is driven primarily by the perceived speed and complexity of AI change rather than by interaction with specific tools. Affective-anxiety items also loaded strongly on their latent factor (λ\u0026thinsp;=\u0026thinsp;0.692\u0026ndash;0.787), particularly those capturing fears of technological dependence and labour displacement, including \u0026ldquo;I fear AI could replace human workers\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.749), \u0026ldquo;I fear humanoid robots will take jobs\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.772), and \u0026ldquo;I fear becoming too dependent on AI technologies\u0026rdquo; (λ\u0026thinsp;=\u0026thinsp;0.787). These results suggest that emotional unease surrounding AI is closely linked to concerns about autonomy and future employment. Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e displays the standardized factor loadings and item-level statistics for the cognitive and affective dimensions of the Albanian AI Anxiety Scale (AIAS). \u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt; Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e here \u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eConfirmatory Factor Loadings for the Albanian AIAS Scale\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eItem (indicator)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEstimate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ez\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e95% CI Lower\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e95% CI Upper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eStandardized Estimate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eSD\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAI Cognitive Anxiety\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLearning to use AI products makes me anxious.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.711\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.857\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.768\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.03\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnderstanding AI functions feels overwhelming.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.733\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.035\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.664\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.802\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.732\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReading an AI manual makes me nervous.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.689\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.033\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.624\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.754\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI feel stressed when I cannot follow AI advances.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.811\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.038\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.737\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.885\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI courses would make me feel anxious.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.753\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.686\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.820\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.770\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInteracting with AI products makes me uneasy.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.729\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.666\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.792\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.751\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.08\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLearning about AI functions increases my anxiety.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.795\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.725\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.865\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.776\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.06\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBeing unable to keep up with AI makes me anxious.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.763\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.696\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.830\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.758\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAI Affective Anxiety\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI fear AI could make me dependent.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.712\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.033\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.647\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.777\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.726\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.09\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI fear AI will make me lazier.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.671\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.611\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.731\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.692\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI fear AI could replace human workers.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.764\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.693\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.835\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.749\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI fear humanoid robots will take jobs.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.801\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.729\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.873\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.772\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.08\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI fear becoming too dependent on AI technologies.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.828\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.039\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.752\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.904\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.787\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI fear AI will replace people\u0026rsquo;s work.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.689\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.623\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.755\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.706\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.12\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eInternal consistency was excellent for both subscales (Cognitive Anxiety α\u0026thinsp;=\u0026thinsp;0.91; Affective Anxiety α\u0026thinsp;=\u0026thinsp;0.88). The latent factors were moderately correlated (r\u0026thinsp;=\u0026thinsp;0.52, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), indicating that while cognitive and affective anxiety are related, they represent empirically distinct dimensions. This distinction is substantively important, as both dimensions were entered simultaneously into the predictive models and demonstrated differential effects on positive and negative attitudes toward AI.\u003c/p\u003e \u003cp\u003eTaken together, the CFA results confirm that the AIAS retains its original factorial structure in the Albanian higher-education context, with strong construct validity and reliability. Alongside the GAAIS and AT-EAI, the AIAS provides a robust measure of the emotional component of AI attitudes, supporting the joint use of all three instruments in subsequent predictive analyses\u0026mdash;the primary objective of Study 1.\u003c/p\u003e \u003c/div\u003e"},{"header":"6. Prediction of GAAIS by Ethical Attitudes toward AI (AT-EAI) and Anxiety (AIAS) towards AI measures","content":"\u003cp\u003eThis second stage addressed RQ2: To what extent do ethical attitudes toward AI (AT-EAI) and AI-related anxiety (AIAS) explain variation in students\u0026rsquo; positive and negative attitudes toward AI (GAAIS), over and above demographic factors? We first assessed concurrent validity through zero-order associations between GAAIS and the ethical/anxiety dimensions (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). We then estimated two hierarchical multiple regression models predicting Positive GAAIS and Negative GAAIS (Tables\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e and \u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e). This approach evaluates predictive validity, i.e., whether ethical cognition and AI anxiety explain AI attitudes in theoretically expected directions once sociodemographic heterogeneity is controlled for.\u003c/p\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e6.1. Concurrent validity: zero-order associations\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e shows a coherent pattern of associations between the GAAIS dimensions and the ethical (AT-EAI) and anxiety (AIAS) components. For Positive GAAIS, correlations with the five AT-EAI subscales were generally small and positive. Fairness (r\u0026thinsp;=\u0026thinsp;0.043, p\u0026thinsp;=\u0026thinsp;0.31) and transparency (r\u0026thinsp;=\u0026thinsp;0.063, p\u0026thinsp;=\u0026thinsp;0.12) were positive but non-significant, whereas non-maleficence showed a small but significant association (r\u0026thinsp;=\u0026thinsp;0.117, p\u0026thinsp;=\u0026thinsp;0.004). This suggests that, in this context, favourable orientations toward AI are most clearly supported by the belief that AI should avoid harm, rather than by fairness or transparency alone. \u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt; Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e here \u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAssociations between the General Attitudes toward Artificial Intelligence Scale and the Ethical Attitudes toward AI (AT-EAI) and AI Anxiety Components (AIAS)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFairness\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTransparency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNon-maleficence\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePrivacy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eResponsibility\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCognitive Anxiety\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eAffective Anxiety\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive GAAIS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003er\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.043\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.063\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.117\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.012\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.141\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.083\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eNegative GAAIS\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003er\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.107\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.154\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.184\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.146\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.407\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e-0.455\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe anxiety components were also significantly related to positive attitudes, though with a mixed pattern: cognitive anxiety correlated negatively with Positive GAAIS (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.141, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while affective anxiety correlated positively but weakly (r\u0026thinsp;=\u0026thinsp;0.083, p\u0026thinsp;=\u0026thinsp;0.04). This indicates that students who feel overwhelmed by AI developments tend to be less optimistic, whereas mild affective concern does not necessarily eliminate enthusiasm for AI\u0026rsquo;s potential benefits. For Negative GAAIS, associations were stronger and more systematic. Ethical dimensions linked to safeguards and rights\u0026mdash;especially privacy (r\u0026thinsp;=\u0026thinsp;0.184, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and responsibility (r\u0026thinsp;=\u0026thinsp;0.146, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001)\u0026mdash;were positively related to negative attitudes, indicating that students who place greater emphasis on data protection and accountability are also more alert to AI risks. Conversely, fairness (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.107, p\u0026thinsp;=\u0026thinsp;0.007) and transparency (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.154, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) were negatively associated with negative attitudes, suggesting that believing AI can be designed to be fair and explainable corresponds to lower levels of AI \u0026ldquo;demonisation.\u0026rdquo;\u003c/p\u003e \u003cp\u003eThe strongest concurrent associations, however, were observed for anxiety: both cognitive anxiety (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.407, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and affective anxiety (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.455, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) showed large associations with negative attitudes, consistent with the view that anxiety functions as a primary affective driver of perceived AI threat. Overall, the correlation matrix supports concurrent validity: the GAAIS dimensions relate to ethical cognition and anxiety in theoretically meaningful ways, but ethical dimensions are not equally salient, and anxiety exhibits the largest associations \u0026ndash; especially with negative AI attitudes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e6.2. Predicting positive GAAIS: the role of non-maleficence and anxiety\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e presents the full hierarchical model predicting Positive GAAIS, which explains a substantial share of variance (R\u0026sup2; = 0.472; F(20, 684)\u0026thinsp;=\u0026thinsp;7.08; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e \u003cp\u003eIn Block 1 (demographics), two results stand out. First, male students reported significantly more positive attitudes than female students (B\u0026thinsp;=\u0026thinsp;0.215, β\u0026thinsp;=\u0026thinsp;0.11, p\u0026thinsp;=\u0026thinsp;0.016). Second, self-reported AI information was negatively associated with Positive GAAIS: compared with those reporting \u0026ldquo;no information,\u0026rdquo; students reporting \u0026ldquo;little\u0026rdquo; (B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.293, p\u0026thinsp;=\u0026thinsp;0.028), \u0026ldquo;enough\u0026rdquo; (B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.568, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and especially \u0026ldquo;much\u0026rdquo; AI information (B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;1.127, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) had lower positive attitude scores.\u003c/p\u003e \u003cp\u003e\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt; Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e here \u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCoefficients for the Positive GAAIS full model (Model 3)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eB\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003et\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e95% CI B lower\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003e95% CI B upper\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Constant)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.645\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.296\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.030\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.064\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.227\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eDemographics\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.215\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.090\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.040\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.391\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge 18\u0026ndash;25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.096\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.130\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.461\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.159\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.351\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge 26\u0026ndash;35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.069\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.117\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;0.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.559\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.299\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.162\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBachelor student\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.394\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.236\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.095\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.857\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.069\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBachelor degree\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.569\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.235\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;2.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;1.030\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u0026minus;0.108\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaster student\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.471\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.241\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.051\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.943\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaster degree\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.361\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.219\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.791\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.069\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLittle AI information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.293\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.133\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;2.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.555\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u0026minus;0.031\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEnough AI information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.568\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.141\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;4.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.846\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u0026minus;0.291\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMuch AI information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;1.127\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.265\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;4.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;1.648\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u0026minus;0.606\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAT-EAI dimensions\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFairness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.017\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.651\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.090\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTransparency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.055\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.040\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.169\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.134\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNon-maleficence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.125\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.054\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.196\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrivacy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;0.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.711\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.084\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.058\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResponsibility\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.019\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.039\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.625\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.057\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.095\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAI Anxiety components\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCognitive anxiety\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.164\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;6.56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.213\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u0026minus;0.115\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAffective anxiety\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.191\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.026\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;7.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.243\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e\u0026minus;0.139\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAlthough counter-intuitive, this pattern plausibly reflects that greater exposure is accompanied by greater awareness of limitations, risks, and contested uses (e.g., generative AI in assessment contexts, proctoring, or plagiarism detection), thereby reducing uncritical enthusiasm. In Block 2 (ethical attitudes), only non-maleficence significantly predicted higher Positive GAAIS (B\u0026thinsp;=\u0026thinsp;0.125, β\u0026thinsp;=\u0026thinsp;0.17, p\u0026thinsp;=\u0026thinsp;0.001). Fairness, transparency, privacy, and responsibility were non-significant. This reinforces the bivariate pattern: safety and harm-avoidance are the ethical dimensions most strongly tied to favourable AI orientations in this sample. In Block 3 (AI anxiety), both anxiety dimensions emerged as strong negative predictors: cognitive anxiety (B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.164, β = \u0026minus;0.25, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and affective anxiety (B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.191, β = \u0026minus;0.29, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Holding demographics and ethical cognition constant, anxiety appears to be the most powerful suppressor of positive AI attitudes. Students who feel unable to keep pace with AI developments or who fear dependence and displacement report markedly lower optimism, even when they endorse ethical AI principles\u0026mdash;highlighting the centrality of the affective pathway in attitude formation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e6.3. Predicting negative GAAIS: anxiety and privacy as drivers of AI risk salience\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e presents the parallel model predicting Negative GAAIS, which is also statistically significant (R\u0026sup2; = 0.392; F(20, 684)\u0026thinsp;=\u0026thinsp;6.88; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), though with slightly lower explanatory power than the positive-attitude model.\u003c/p\u003e \u003cp\u003e\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt; Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e here \u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCoefficients for the Negative GAAIS full model (Model 3)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eB\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003et\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e95% CI B lower\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003e95% CI B upper\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Constant)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.041\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.296\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.890\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.541\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.622\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eDemographics\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale (1\u0026thinsp;=\u0026thinsp;male)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.332\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.090\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.156\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.507\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge 18\u0026ndash;25 (ref\u0026thinsp;=\u0026thinsp;36\u0026ndash;50)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.215\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.130\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.099\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.040\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.471\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge 26\u0026ndash;35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.227\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.117\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.054\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.457\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBachelor student\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.383\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.236\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.105\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.847\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.080\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBachelor degree\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.398\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.235\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.091\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.859\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.064\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaster student\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.196\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.241\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;0.81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.417\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.668\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.277\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaster degree\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.136\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.219\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.534\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.567\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.294\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLittle AI information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.131\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.133\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;0.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.327\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.393\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.131\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEnough AI information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.231\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.141\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.102\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.509\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.046\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMuch AI information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.266\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.917\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.549\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.494\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAT-EAI dimensions\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFairness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.053\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.159\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.126\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTransparency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.040\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.066\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.153\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNon-maleficence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;0.051\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.161\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.122\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.020\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrivacy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.092\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.163\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResponsibility\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.039\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.343\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.040\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.113\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAI Anxiety components\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCognitive anxiety\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.151\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.030\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.092\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.210\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAffective anxiety\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.167\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.106\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.228\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAmong demographic controls, male students again scored higher (B\u0026thinsp;=\u0026thinsp;0.332, β\u0026thinsp;=\u0026thinsp;0.17, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Taken together with the positive-attitude findings, this indicates a more polarised or salient attitudinal profile among male students \u0026ndash; greater endorsement of both promise and risk \u0026ndash; rather than a simple optimism\u0026ndash;pessimism shift. Age and education were not statistically significant in the full model.\u003c/p\u003e \u003cp\u003eAmong ethical attitudes, only privacy predicted higher negative attitudes (B\u0026thinsp;=\u0026thinsp;0.092, β\u0026thinsp;=\u0026thinsp;0.10, p\u0026thinsp;=\u0026thinsp;0.011). Students who place greater importance on consent, data protection, and control over personal information are more likely to endorse risk-oriented statements. Fairness and non-maleficence had negative coefficients, as expected, but were not significant once anxiety was included, suggesting that ethical reassurance may partly operate through emotional responses or be overshadowed by affective threat perceptions.\u003c/p\u003e \u003cp\u003eAs hypothesised, AI anxiety was the dominant predictor of negative GAAIS. Both cognitive anxiety (B\u0026thinsp;=\u0026thinsp;0.151, β\u0026thinsp;=\u0026thinsp;0.29, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and affective anxiety (B\u0026thinsp;=\u0026thinsp;0.167, β\u0026thinsp;=\u0026thinsp;0.32, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) were strong and highly significant. Standardised coefficients indicate that affective anxiety was the single strongest predictor (β\u0026thinsp;=\u0026thinsp;0.32), slightly exceeding cognitive anxiety (β\u0026thinsp;=\u0026thinsp;0.29), implying that fear-based concerns (dependency and displacement) are especially influential in elevating perceived AI risk.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e6.4. Concurrent and predictive validity\u003c/h2\u003e \u003cp\u003eTogether, results provide strong evidence of both concurrent and predictive validity for the Albanian adaptations of the AT-EAI, AIAS, and GAAIS. Ethical and affective predictors jointly explain substantial variance in attitudes after controlling for demographic differences (R\u0026sup2; = 0.47 for Positive GAAIS; R\u0026sup2; = 0.39 for Negative GAAIS). Across models, two patterns are especially robust.\u003c/p\u003e \u003cp\u003eFirst, among ethical dimensions, non-maleficence is the key facilitator of positive attitudes, while privacy is the key amplifier of negative attitudes\u0026mdash;suggesting that safety and data protection are the most consequential ethical \u0026ldquo;entry points\u0026rdquo; shaping students\u0026rsquo; evaluations of AI in higher education. Second, both forms of AI anxiety are consistently strong predictors: cognitive anxiety and affective anxiety suppress positive attitudes and intensify negative attitudes, with affective anxiety showing the largest effect on risk salience.\u003c/p\u003e \u003cp\u003eGender differences are noteworthy: male students reported both stronger positive and stronger negative attitudes, consistent with a more polarised attitude profile. Finally, greater self-reported AI information predicted lower positive attitudes, suggesting that increased familiarity may be accompanied by more critical awareness of limitations and contested uses of AI tools in academic settings. Although counter-intuitive, this pattern is consistent with an \u0026lsquo;informed scepticism\u0026rsquo; mechanism: greater exposure may increase awareness of contested uses (e.g., surveillance-like proctoring, opaque plagiarism detection, dataset bias, and academic-integrity dilemmas), reducing uncritical enthusiasm. In contexts where institutional guidance is fragmented, information may be acquired through social media narratives and peer experience, which can amplify critical frames rather than build calibrated trust. Table\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e summarizes the research hypotheses alongside the observed results from the predictive models, highlighting areas of support and divergence. \u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt;\u0026gt; Table\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e here \u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u0026lt;\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary of Hypotheses and Results\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePredictor\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHypothesized Direction\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePositive GAAIS (Observed)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHypothesized Direction\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNegative GAAIS (Observed)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFairness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTransparency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNegative (marginal, p\u0026thinsp;\u0026asymp;\u0026thinsp;.07)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNon-maleficence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePositive (p\u0026thinsp;\u0026lt;\u0026thinsp;.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrivacy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePositive (p\u0026thinsp;\u0026lt;\u0026thinsp;.05)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResponsibility\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI Anxiety (Cognitive)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNegative (p\u0026thinsp;\u0026lt;\u0026thinsp;.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePositive (p\u0026thinsp;\u0026lt;\u0026thinsp;.001) \u0026rarr; higher anxiety \u0026rarr; greater risk salience\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI Anxiety (Affective)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNegative (p\u0026thinsp;\u0026lt;\u0026thinsp;.001) (weak positive bivariate p\u0026thinsp;\u0026lt;\u0026thinsp;.05)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePositive (p\u0026thinsp;\u0026lt;\u0026thinsp;.001) (strongest predictor)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender (1\u0026thinsp;=\u0026thinsp;Male)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003en/a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePositive (p\u0026thinsp;\u0026lt;\u0026thinsp;.05)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003en/a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePositive (p\u0026thinsp;\u0026lt;\u0026thinsp;.001)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003en/a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003en/a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation level\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNegative (p\u0026thinsp;\u0026lt;\u0026thinsp;.10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI Information level\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNegative (p\u0026thinsp;\u0026lt;\u0026thinsp;.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003en.s.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFrom a policy and practice perspective, the findings imply that ethical reassurance alone is insufficient to foster favourable attitudes unless anxiety is simultaneously addressed and visible safeguards are in place. For Albanian higher education, AI literacy efforts may be most effective when they integrate technical competence, ethical reflection, and affective preparedness, with particular emphasis on \u0026ldquo;do no harm\u0026rdquo; principles and data protection by design in AI-enabled teaching and assessment systems.\u003c/p\u003e \u003c/div\u003e"},{"header":"7. Discussion","content":"\u003cp\u003eThis study examined how ethical cognition and emotional responses jointly shape university students\u0026rsquo; general attitudes toward artificial intelligence (AI) in Albania. Drawing on validated Albanian versions of the GAAIS, AT-EAI, and AIAS, the findings show that students\u0026rsquo; evaluations of AI are structurally ambivalent: optimism about AI\u0026rsquo;s potential coexists with salient apprehension. This pattern is consistent with contemporary views of technology attitudes as multi-dimensional rather than a single acceptance\u0026ndash;rejection continuum.\u003c/p\u003e \u003cp\u003eThe results support affective\u0026ndash;cognitive models of attitude formation (Ajzen \u0026amp; Fishbein; Eagly \u0026amp; Chaiken) by demonstrating that ethical evaluations and emotional reactions are empirically distinguishable and jointly relevant. Importantly, not all ethical concerns mattered equally. Non-maleficence (\u0026ldquo;do no harm\u0026rdquo;) was the only ethical dimension that reliably supported more positive attitudes, suggesting that safety functions as the primary moral \u0026ldquo;threshold condition\u0026rdquo; for endorsing AI in higher education. By contrast, privacy concerns amplified negative attitudes, indicating that data protection and control over personal information act as a rights-based lens through which AI risks become more salient\u0026mdash;particularly in educational environments where monitoring, assessment, and data processing are central.\u003c/p\u003e \u003cp\u003eAI anxiety emerged as the strongest and most consistent predictor across outcomes. Both cognitive anxiety (difficulty understanding or keeping up with AI) and affective anxiety (fear of dependence or displacement) reduced positive attitudes and increased negative attitudes, with affective anxiety showing the largest association with risk-oriented evaluations. This aligns with affect heuristic and dual-process accounts of risk perception (Slovic et al.), where emotional cues operate as rapid signals of threat under uncertainty. The moderate association between positive and negative GAAIS dimensions further indicates that endorsement of AI\u0026rsquo;s benefits does not preclude discomfort about its consequences; rather, students appear to evaluate AI through parallel \u0026ldquo;opportunity\u0026rdquo; and \u0026ldquo;risk\u0026rdquo; channels.\u003c/p\u003e \u003cp\u003eThese findings also extend technology acceptance perspectives (TAM/UTAUT2) by showing that instrumental familiarity is not sufficient to explain AI attitudes in higher education. Once ethical and affective factors are considered, demographics play a comparatively smaller role, and greater self-reported AI knowledge is associated with lower positivity. A plausible interpretation is that increased exposure brings more critical awareness of contested academic uses (e.g., integrity systems, surveillance-like monitoring, opaque decisions), reducing unreflective enthusiasm. In this sense, AI literacy may produce \u0026ldquo;informed scepticism\u0026rdquo; as well as competence, suggesting that the knowledge\u0026ndash;acceptance relationship is not necessarily linear.\u003c/p\u003e \u003cp\u003eThe Albanian context helps interpret why non-maleficence and privacy are especially salient. Students\u0026rsquo; encounters with AI often occur through externally introduced systems\u0026mdash;plagiarism detection, automated evaluation, and generative AI tools\u0026mdash;within an institutional setting where governance standards and ethical guidance are still consolidating. Under such conditions, safety and data protection become practical proxies for trust: when formal assurances are limited, students rely more strongly on moral safeguards and on their emotional sense of control.\u003c/p\u003e \u003cp\u003eFrom a higher-education perspective, the results imply that responsible AI integration should not be framed as a purely technical upgrade. Universities are likely to see better acceptance when AI literacy is paired with (i) visible harm-avoidance safeguards and (ii) clear data-governance practices that make privacy protections concrete. Equally, anxiety is not a peripheral attitude \u0026ldquo;noise\u0026rdquo; but a core mechanism shaping receptiveness; reducing anxiety requires structured opportunities for guided exposure, critical discussion, and transparent explanation of how AI-enabled systems work and how students are protected.\u003c/p\u003e \u003cp\u003eSeveral limitations should be noted. The cross-sectional design limits causal inference, and reliance on self-report measures may inflate associations through shared method variance. Future work should use longitudinal or experimental designs (e.g., pre/post AI-ethics training) to assess whether anxiety decreases and attitudes shift with structured exposure and clearer institutional safeguards. Extending the model to include perceived control, institutional trust, and perceived fairness of university AI practices would also help clarify the pathways through which ethics and anxiety translate into positive and negative evaluations.\u003c/p\u003e \u003cp\u003eThis study offers compelling evidence that attitudes toward artificial intelligence in Albanian higher education are shaped by an interplay of ethical cognition and affective responses. The validated Albanian versions of the AT-EAI, AIAS, and GAAIS demonstrated excellent psychometric properties, enabling a rigorous examination of how moral evaluations\u0026mdash;particularly non-maleficence and privacy\u0026mdash;and emotional factors jointly predict both positive and negative AI attitudes. Ethical endorsement of non-maleficence emerged as the most consistent driver of favourable evaluations, whereas privacy concerns heightened sensitivity to perceived risks.\u003c/p\u003e \u003cp\u003eAffective and cognitive facets of anxiety proved especially consequential: both significantly suppressed positive attitudes and intensified negative ones, with affective anxiety exerting the strongest influence on risk perception. These findings reinforce the centrality of emotion within technology-acceptance processes and point to broader dynamics characteristic of post-transition institutional contexts, where ethical reassurance often compensates for limited regulatory clarity or institutional trust.\u003c/p\u003e \u003cp\u003eTheoretically, the study integrates cognitive, affective, and ethical dimensions into a unified framework of AI acceptance, demonstrating that attitudes toward AI extend beyond utilitarian appraisals to encompass deeper moral and emotional layers. Despite its contributions, the study\u0026rsquo;s cross-sectional design and reliance on self-report measures constrain causal inference and raise the possibility of common-method bias. Future work should therefore adopt longitudinal, experimental, and cross-cultural designs to further refine and generalize the proposed model.\u003c/p\u003e \u003cp\u003eTo strengthen students\u0026rsquo; engagement with artificial intelligence, higher education institutions should integrate both ethical and emotional dimensions into AI literacy programs, ensuring that learners develop critical understanding and reduced anxiety. Institutional trust must also be reinforced through transparent data policies, clear accountability mechanisms, and human oversight, which are essential in transitional governance contexts. Universities should implement targeted, evidence-based interventions\u0026mdash;such as workshops, reflective discussions, and counselling services\u0026mdash;to address cognitive and affective dimensions of AI-related anxiety. Future research should examine how contextual and cultural factors shape AI perceptions, with particular attention to the role of institutional trust. Finally, adopting longitudinal and experimental research designs would improve causal inference and track how ethical cognition and anxiety evolve over time.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was conducted in accordance with the ethical standards of the University of Shkodra “Luigj Gurakuqi” and with the principles of the Declaration of Helsinki. Ethical approval was obtained from the University Ethics Committee of the University of Shkodra “Luigj Gurakuqi” (Approval No. 2025/AEI-05). All participants were informed about the purpose of the study, the voluntary nature of participation, and their right to withdraw at any time without consequence. Written informed consent was obtained from all participants prior to data collection.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable. The manuscript does not contain any individual person’s data in any identifiable form.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated and/or analysed during the current study are not publicly available due to ethical and privacy considerations related to the protection of participants’ anonymity, but are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors’ contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eE.D. conceived the study, designed the research framework, performed the statistical analyses, and drafted the manuscript.\u003c/p\u003e\n\u003cp\u003eE.H. translated and culturally adapted the instruments, conducted pilot testing, coordinated data collection, and contributed to data preparation and validation.\u003c/p\u003e\n\u003cp\u003eBoth authors reviewed, revised, and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors wish to thank the students and academic staff of the University of Shkodra “Luigj Gurakuqi” for their participation and valuable insights. We are also grateful to the University Ethics Committee for their guidance and approval of the research protocol. Special thanks to colleagues from the Faculty of Economy and the Faculty of Educational Sciences for their feedback during the instrument adaptation and piloting phases. Finally, we acknowledge the support of the university’s administrative and IT staff in facilitating both online and in-class data collection.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors’ information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eE.D. is a lecturer in research methods and applied economics at the department of Business Administration, Faculty of Economy, University of Shkodra “Luigj Gurakuqi”, with research interests in labour markets, inequality, technology acceptance, and the social and ethical implications of artificial intelligence.\u003c/p\u003e\n\u003cp\u003eE.H. is a lecturer in psychology and serves as the head of the Social Work and Psychology Department, Faculty of Educational Sciences, the University of Shkodra “Luigj Gurakuqi”, with research interests in educational psychology, measurement, and digital transformation in education.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAjzen, I., 1982. Equity in Attitude Formation and Change, in: Greenberg, J., Cohen, R.L. (Eds.), Equity and Justice in Social Behavior. Academic Press, pp. 161\u0026ndash;186. https://doi.org/10.1016/B978-0-12-299580-4.50011-2\u003c/li\u003e\n\u003cli\u003eAjzen, I., Fishbein, M., 2000. Attitudes and the Attitude-Behavior Relation: Reasoned and Automatic Processes. European Review of Social Psychology 11, 1\u0026ndash;33. https://doi.org/10.1080/14792779943000116\u003c/li\u003e\n\u003cli\u003eArora, N., Garg, N., 2023. Meaningful Work in the Digital Age- A Comprehensive Review and Framework. Human Resource Development International 0, 1\u0026ndash;25. https://doi.org/10.1080/13678868.2024.2336866\u003c/li\u003e\n\u003cli\u003eBankins, S., Formosa, P., 2023. The Ethical Implications of Artificial Intelligence (AI) For Meaningful Work. J Bus Ethics 185, 725\u0026ndash;740. https://doi.org/10.1007/s10551-023-05339-7\u003c/li\u003e\n\u003cli\u003eCicero, L., Russo, A., Di Stefano, G., Zammitti, A., 2025. The General Attitudes towards Artificial Intelligence Scale (GAAIS): validation and psychometric properties analysis in the Italian context. BMC Psychol 13, 641. https://doi.org/10.1186/s40359-025-02935-2\u003c/li\u003e\n\u003cli\u003eCouncil of Ministers, 2025. Albania 2030 \u0026ndash; Technology as a Pillar of Integration into the European Union. Albanian Government Council of Ministers. URL https://www.kryeministria.al/en/newsroom/shqiperia-2030-teknologjia-si-shtylle-e-integrimit-ne-bashkimin-europian/ (accessed 12.21.25).\u003c/li\u003e\n\u003cli\u003eDavis, F.D., 1989. Technology acceptance model: TAM. Al-Suqri, MN, Al-Aufi, AS: Information Seeking Behavior and Technology Adoption 205.\u003c/li\u003e\n\u003cli\u003eEagly, A.H., Chaiken, S., 1993. The psychology of attitudes, The psychology of attitudes. Harcourt Brace Jovanovich College Publishers, Orlando, FL, US.\u003c/li\u003e\n\u003cli\u003eETF, 2022. Embracing the digital age | ETF.\u003c/li\u003e\n\u003cli\u003eEU4Youth, 2024. Youth Guarantee in Albania: Real opportunities for employment, skills and education. | UNICEF Albania. URL https://www.unicef.org/albania/stories/youth-guarantee-albania-real-opportunities-employment-skills-and-education (accessed 7.16.25).\u003c/li\u003e\n\u003cli\u003eEuropean Commission, 2025a. 2025 Communication on EU enlargement policy.\u003c/li\u003e\n\u003cli\u003eEuropean Commission, 2025b. The Commission publishes guidelines on AI system definition to facilitate the first AI Act\u0026rsquo;s rules application | Shaping Europe\u0026rsquo;s digital future.\u003c/li\u003e\n\u003cli\u003eGoA, 2025. Per miratimin e Strategjise Kombetare te Inteligjences Artificiale te Shqiperise 2025-2030 dhe te Planit te Veprimit. Council of Ministers, Tirana, Albania.\u003c/li\u003e\n\u003cli\u003eHubertz, M., Janowsky, A., 2025. Mind the Gap: Understanding Diverse Student Experiences with AI Through GAAIS Analysis, in: Smith, B.K., Borge, M. (Eds.), Learning and Collaboration Technologies. Springer Nature Switzerland, Cham, pp. 252\u0026ndash;269. https://doi.org/10.1007/978-3-031-93746-0_18\u003c/li\u003e\n\u003cli\u003eINSTAT, 2024. Arsimi - Shifrat.\u003c/li\u003e\n\u003cli\u003eJang, Y., Choi, S., Kim, H., 2022. Development and validation of an instrument to measure undergraduate students\u0026rsquo; attitudes toward the ethics of artificial intelligence (AT-EAI) and analysis of its difference by gender and experience of AI education. Educ Inf Technol 27, 11635\u0026ndash;11667. https://doi.org/10.1007/s10639-022-11086-5\u003c/li\u003e\n\u003cli\u003eJobin, A., Ienca, M., Vayena, E., 2019. The global landscape of AI ethics guidelines. Nat Mach Intell 1, 389\u0026ndash;399. https://doi.org/10.1038/s42256-019-0088-2\u003c/li\u003e\n\u003cli\u003eKaya, F., Aydin, F., Schepman, A., Rodway, P., Yetişensoy, O., Demir Kaya, M., 2024. The Roles of Personality Traits, AI Anxiety, and Demographic Factors in Attitudes toward Artificial Intelligence. International Journal of Human\u0026ndash;Computer Interaction 40, 497\u0026ndash;514. https://doi.org/10.1080/10447318.2022.2151730\u003c/li\u003e\n\u003cli\u003eKong, S.-C., Cheung, W.M.-Y., Tsang, O., 2023. Evaluating an artificial intelligence literacy programme for empowering and developing concepts, literacy and ethical awareness in senior secondary students. Educ Inf Technol 28, 4703\u0026ndash;4724. https://doi.org/10.1007/s10639-022-11408-7\u003c/li\u003e\n\u003cli\u003eLee, M.K., Kusbit, D., Metsky, E., Dabbish, L., 2015. Working with Machines: The Impact of Algorithmic and Data-Driven Management on Human Workers, in: Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems. Presented at the CHI \u0026rsquo;15: CHI Conference on Human Factors in Computing Systems, ACM, Seoul Republic of Korea, pp. 1603\u0026ndash;1612. https://doi.org/10.1145/2702123.2702548\u003c/li\u003e\n\u003cli\u003eLim, T., Gottipati, S., Cheong, M., 2025. What students really think: unpacking AI ethics in educational assessments through a triadic framework. Int J Educ Technol High Educ 22, 56. https://doi.org/10.1186/s41239-025-00556-8\u003c/li\u003e\n\u003cli\u003eLiska, A.E., 1984. A Critical Examination of the Causal Structure of the Fishbein/Ajzen Attitude-Behavior Model. Social Psychology Quarterly 47, 61\u0026ndash;74. https://doi.org/10.2307/3033889\u003c/li\u003e\n\u003cli\u003eLongoni, C., Cian, L., 2022. Artificial Intelligence in Utilitarian vs. Hedonic Contexts: The \u0026ldquo;Word-of-Machine\u0026rdquo; Effect. Journal of Marketing 86, 91\u0026ndash;108. https://doi.org/10.1177/0022242920957347\u003c/li\u003e\n\u003cli\u003eMantello, P.A., Ghotbi, N., Ho, M.-T., Mizutani, F., 2025. Gauging public opinion of AI and emotionalized AI in healthcare: findings from a nationwide survey in Japan. AI \u0026amp; Soc 40, 3735\u0026ndash;3749. https://doi.org/10.1007/s00146-024-02126-4\u003c/li\u003e\n\u003cli\u003eMAS, 2025. Dokumenti i Politikave te Arsimit te Larte. Ministria e Arsimit dhe Sportit.\u003c/li\u003e\n\u003cli\u003eMASR, 2021. Draft-Strategjia Komb\u0026euml;tare e Arsimit 2021-2026. Ministria e Arsimit, Sportit dhe Rinis\u0026euml;.\u003c/li\u003e\n\u003cli\u003eMubashir, A.S., Altaf, W., Kainaat, R., 2025. Personality Traits and Attitudes Towards Artificial Intelligence Among University Students. Social Science Review Archives 3, 1010\u0026ndash;1019. https://doi.org/10.70670/sra.v3i2.690\u003c/li\u003e\n\u003cli\u003eNomura, T., Tanaka, M., 2022. Experiences, knowledge of functions, and social acceptance of robots: an exploratory case study focusing on Japan. AI \u0026amp; Soc 37, 367\u0026ndash;374. https://doi.org/10.1007/s00146-021-01196-y\u003c/li\u003e\n\u003cli\u003eSacco, F., Manzi, F., Di Dio, C., Schepman, A., Rodway, P., Massaro, D., Marchetti, A., 2025. Validation of the general attitudes towards Artificial Intelligence Scale in the Italian context and the role of trust. Computers in Human Behavior Reports 19, 100751. https://doi.org/10.1016/j.chbr.2025.100751\u003c/li\u003e\n\u003cli\u003eSchepman, A., Rodway, P., 2025. The Measurement of Attitudes Towards Artificial Intelligence: An Overview and Recommendations, in: Montag, C., Ali, R. (Eds.), The Impact of Artificial Intelligence on Societies: Understanding Attitude Formation Towards AI. Springer Nature Switzerland, Cham, pp. 9\u0026ndash;24. https://doi.org/10.1007/978-3-031-70355-3_2\u003c/li\u003e\n\u003cli\u003eSchepman, A., Rodway, P., 2020. Initial validation of the general attitudes towards Artificial Intelligence Scale. Computers in Human Behavior Reports 1, 100014. https://doi.org/10.1016/j.chbr.2020.100014\u003c/li\u003e\n\u003cli\u003eSkagerlund, K., Forsblad, M., Slovic, P., V\u0026auml;stfj\u0026auml;ll, D., 2020. The Affect Heuristic and Risk Perception \u0026ndash; Stability Across Elicitation Methods and Individual Cognitive Abilities. Front. Psychol. 11. https://doi.org/10.3389/fpsyg.2020.00970\u003c/li\u003e\n\u003cli\u003eSlovic, P., Finucane, M.L., Peters, E., MacGregor, D.G., 2007. The affect heuristic. European Journal of Operational Research 177, 1333\u0026ndash;1352. https://doi.org/10.1016/j.ejor.2005.04.006\u003c/li\u003e\n\u003cli\u003eSlovic, P., Finucane, M.L., Peters, E., MacGregor, D.G., 2004. Risk as analysis and risk as feelings: some thoughts about affect, reason, risk, and rationality. Risk Anal 24, 311\u0026ndash;322. https://doi.org/10.1111/j.0272-4332.2004.00433.x\u003c/li\u003e\n\u003cli\u003eSultana, A., Abdul Latheef, N., Siby, N., Ahmad, Z., 2025. Exploring Students\u0026rsquo; Attitudes Toward Artificial Intelligence (AI): Psychometric Validation of AI-Attitude Scale. Sage Open 15, 21582440251378375. https://doi.org/10.1177/21582440251378375\u003c/li\u003e\n\u003cli\u003eSun, L., Zhou, L., 2025. Generative artificial intelligence attitude analysis of undergraduate students and their precise improvement strategies: A differential analysis of multifactorial influences. Educ Inf Technol 30, 10591\u0026ndash;10626. https://doi.org/10.1007/s10639-024-13236-3\u003c/li\u003e\n\u003cli\u003eVenkatesh, V., Morris, M.G., Davis, G.B., Davis, F.D., 2003. User Acceptance of Information Technology: Toward a Unified View. MIS Quarterly 27, 425\u0026ndash;478. https://doi.org/10.2307/30036540\u003c/li\u003e\n\u003cli\u003eWang, W., Downey, J., Yang, F., 2025. AI anxiety? Comparing the sociotechnical imaginaries of artificial intelligence in UK, Chinese and Indian newspapers. Global Media and China 10, 425\u0026ndash;441. https://doi.org/10.1177/20594364231196547\u003c/li\u003e\n\u003cli\u003eWang, Y.-Y., Wang, Y.-S., 2022. Development and validation of an artificial intelligence anxiety scale: an initial application in predicting motivated learning behavior. Interactive Learning Environments 30, 619\u0026ndash;634. https://doi.org/10.1080/10494820.2019.1674887\u003c/li\u003e\n\u003cli\u003eWeng, X., Xia, Q., Ahmad, Z., Chiu, T.K.F., 2024. Personality traits for self-regulated learning with generative artificial intelligence: The case of ChatGPT. Computers and Education: Artificial Intelligence 7, 100315. https://doi.org/10.1016/j.caeai.2024.100315\u003c/li\u003e\n\u003cli\u003eXue, L., Rashid, A.M., Ouyang, S., 2024. The Unified Theory of Acceptance and Use of Technology (UTAUT) in Higher Education: A Systematic Review. Sage Open 14, 21582440241229570. https://doi.org/10.1177/21582440241229570\u003c/li\u003e\n\u003cli\u003eYilmaz, F.G.K., Yilmaz, R., Ceylan, M., 2024. Generative Artificial Intelligence Acceptance Scale: A Validity and Reliability Study. International Journal of Human\u0026ndash;Computer Interaction 40, 8703\u0026ndash;8715. https://doi.org/10.1080/10447318.2023.2288730\u003c/li\u003e\n\u003cli\u003eZhu, X., Gan, J., Yang, X., 2025. Social mobility beliefs moderate links between SSS, attributions for wealth and poverty, and aggression. BMC Psychol 13, 820. https://doi.org/10.1186/s40359-025-03138-5\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-psychology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"psyo","sideBox":"Learn more about [BMC Psychology](http://bmcpsychology.biomedcentral.com/)","snPcode":"","submissionUrl":"","title":"BMC Psychology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Artificial intelligence, higher education, Albania, ethical cognition, AI anxiety, technology acceptance","lastPublishedDoi":"10.21203/rs.3.rs-8483126/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8483126/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eArtificial intelligence is increasingly embedded in higher education, shaping teaching, assessment, and institutional governance. While existing research often focuses on perceived usefulness and performance, less attention has been paid to how ethical evaluations and emotional responses jointly shape attitudes toward artificial intelligence. This study examines attitudes toward artificial intelligence in Albanian higher education through an integrated cognitive\u0026ndash;affective\u0026ndash;ethical framework, focusing on ethical cognition and artificial intelligence\u0026ndash;related anxiety as key determinants of positive and negative attitudes.\u003c/p\u003e \u003cp\u003eA cross-sectional survey was conducted among 705 students and academic staff at Albanian universities. Three validated instruments were translated and culturally adapted: the General Attitudes toward Artificial Intelligence Scale, the Attitudes toward Ethical Artificial Intelligence Scale, and the Artificial Intelligence Anxiety Scale. Confirmatory factor analyses were performed to assess factorial validity and reliability. Hierarchical multiple regression models were then estimated to examine the predictive roles of ethical attitudes and artificial intelligence\u0026ndash;related anxiety on positive and negative attitudes, controlling for sociodemographic characteristics.\u003c/p\u003e \u003cp\u003eConfirmatory factor analyses supported the original factorial structures of all three instruments, with excellent model fit and strong internal consistency. Ethical cognition and emotional responses jointly explained substantial variance in attitudes toward artificial intelligence. Among ethical dimensions, non-maleficence was the only factor consistently associated with more positive attitudes, whereas privacy concerns significantly increased negative evaluations. Both cognitive and affective forms of artificial intelligence anxiety reduced positive attitudes and amplified negative ones, with affective anxiety showing the strongest effects. Ethical and emotional factors together explained 47% of the variance in positive attitudes and 39% in negative attitudes.\u003c/p\u003e \u003cp\u003eAttitudes toward artificial intelligence in higher education are shaped by an interplay of ethical evaluation and emotional response rather than by instrumental considerations alone. Ethical reassurance related to harm avoidance and effective management of anxiety appear central to fostering constructive engagement with artificial intelligence. These findings highlight the importance of integrating ethical reflection and emotional preparedness into artificial intelligence literacy initiatives, particularly in transitional higher-education contexts.\u003c/p\u003e","manuscriptTitle":"Ethical Cognition, Anxiety, and Attitudes toward Artificial Intelligence in Higher Education: Validation and Predictive Modelling of the Albanian GAAIS","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-05 08:27:38","doi":"10.21203/rs.3.rs-8483126/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-19T09:14:23+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-02T10:39:54+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-02T07:01:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"40843629066637952544284197281283495962","date":"2026-01-29T16:16:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"193414735943630300025632687987356607709","date":"2026-01-29T16:00:08+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-29T15:18:08+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-07T17:00:32+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-01T21:57:50+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-01T21:56:52+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Psychology","date":"2025-12-30T16:33:31+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-psychology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"psyo","sideBox":"Learn more about [BMC Psychology](http://bmcpsychology.biomedcentral.com/)","snPcode":"","submissionUrl":"","title":"BMC Psychology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ffbc809c-8327-4cac-a6a2-90a88b35a88f","owner":[],"postedDate":"January 5th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-13T07:56:07+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-05 08:27:38","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8483126","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8483126","identity":"rs-8483126","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}