Artificial Intelligence in Dental Education: A Scoping Review of Opportunities, Challenges, and Ethical Frameworks for Shaping Accreditation Standards and Future Practice

Artificial Intelligence in Dental Education: A Scoping Review of Opportunities, Challenges, and Ethical Frameworks for Shaping Accreditation Standards and Future Practice | 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 Artificial Intelligence in Dental Education: A Scoping Review of Opportunities, Challenges, and Ethical Frameworks for Shaping Accreditation Standards and Future Practice Ayman M. Khalifah, Rasha Alafaleg This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6498214/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: The integration of artificial intelligence (AI) into dental education offers transformative potential for enhancing learning outcomes, clinical training, and institutional efficiency. However, rapid AI adoption introduces ethical, logistical, and pedagogical challenges that require systematic exploration. This scoping review maps the current applications, challenges, and future directions of AI in dental education, focusing on its integration into curricula while ensuring ethical, equitable, and pedagogically sound practices. Methods: The Joanna Briggs Institute framework was followed, with reporting per the PRISMA-ScR guidelines for scoping reviews. A systematic search was conducted across PubMed, EMBASE, MEDLINE-Ovid, and Google Scholar for studies published between January 2018 and January 2025. The search terms included "artificial intelligence," "dental education," "machine learning," "ChatGPT," and "ethical challenges," with Medical Subject Headings (MeSH) terms applied where applicable. After duplicate removal, 624 510 records underwent title/abstract screening, followed by a full-text review of 57 articles, with 43 studies meeting the eligibility criteria. Data extraction focused on the study design, population, AI type, key outcomes, and challenges. Results: The key findings include the following: 1. AI-Driven Personalization: Generative AI (e.g., ChatGPT) reduced grading time by 45% and improved reflective learning outcomes, although 33% of studies reported algorithmic bias due to nonrepresentative training data. 2. In clinical training, AI tools achieved 99% accuracy in caries detection compared with 77–79% accuracy for students, but models trained on homogeneous datasets underperformed in diverse cohorts. 3. Institutional Efficiency : Automated scheduling reduced administrative workloads by 30%, yet only 18% of institutions had updated curricula to include AI literacy modules. 4. Ethical Governance: Data privacy and data protection breaches occurred in 24% of the studies, and 41% reported faculty resistance to AI adoption, highlighting the need for dental-specific guidelines. Conclusion: AI holds significant promise for dental education but requires addressing ethical, logistical, and pedagogical challenges. Future efforts should focus on updating accreditation standards, fostering interdisciplinary collaboration, and developing hybrid models that balance AI-driven efficiency with traditional mentorship. Longitudinal studies are needed to evaluate the long-term impact of AI on clinical competence and patient outcomes. Significance: Dental educators need clearer guidance on integrating AI into the dental curriculum. Artificial intelligence (AI) Dental education Ethical challenges Curriculum design Clinical training Accreditation standards Algorithmic bias Figures Figure 1 Figure 2 Background The integration of artificial intelligence (AI) into healthcare has revolutionized diagnostic accuracy, patient management, and personalized treatment planning [ 1 ]. In education, AI-driven tools are reshaping pedagogical approaches by enabling adaptive learning, automated assessments, and immersive simulations [ 2 ]. Dental education, positioned at the intersection of healthcare and pedagogy, stands to benefit significantly from these advancements. However, the rapid adoption of AI also introduces ethical, logistical, and pedagogical challenges that demand systematic exploration[ 3 ]. The current literature highlights AI’s potential in dental education, such as virtual reality (VR) simulations for procedural training and generative AI for personalized feedback [ 4 , 5 ]. However, existing studies often focus on isolated applications, with a limited synthesis of broader implications, including faculty readiness, curricular alignment, and ethical governance. For example, while AI-enhanced diagnostics improve efficiency, concerns about algorithmic bias and overreliance on automated systems remain understudied [ 6 ]. Furthermore, disparities in AI adoption between high- and low-resource institutions risk exacerbating global inequities in dental training [ 7 ]. Thus, education and training for health professionals and the need for a policy to be identified for AI implementation are areas of importance for future research. Methods This scoping review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) (Appendix 1) [ 8 ]. A comprehensive search strategy was employed across PubMed, EMBASE, MEDLINE-Ovid, and Google Scholar, covering studies published between January 2018 and January 2025. The search terms included combinations of "artificial intelligence," "dental education," "machine learning," "ChatGPT," and "ethical challenges," “with Medical Subject Headings (MeSH) terms applied where applicable. Additionally, manual searches of gray literature, including institutional reports and conference proceedings, were conducted to ensure the broad inclusion of relevant studies. After removing duplicates, 624 records underwent title/abstract screening, followed by a full-text review of 57 articles. Forty-three studies met the eligibility criteria and were included in the final analysis (Fig. 1). Data extraction focused on study design, population, AI type, key outcomes, and challenges, with thematic synthesis categorizing findings into four domains: personalization, clinical training, institutional efficiency, and ethical governance. A scoping review protocol was developed according to the Jonna Briggs Institute framework (JBI) guidelines. Thus, an overview of the literature was mapped to address broader questions. A data extraction (selection and coding) protocol was developed around population/intervention/compare/outcome (PICO). Objective and research question The purpose and objectives of this review are to answer gaps in the literature by systematically mapping AI applications, challenges, and future directions in dental education. The primary research question is as follows: what are the common AI tools used by dental educators at present? How can AI be integrated into dental curricula to enhance learning outcomes while ensuring ethical, equitable, and pedagogically sound practices? This study aims to synthesize evidence on AI’s educational applications, identify barriers to implementation, and propose strategies for responsible integration aligned with global standards. Identifying relevant studies and eligibility criteria For this scoping review to be efficient, it will be important to draw upon the full range of published literature from quantitative and qualitative studies to policies and the editorial press. To include from the academic literature to health services and general literature in all ranges of clinical practice to gather evidence that is relevant to dentistry. Publications in the English language were reviewed as most trade publications and gray literature from 2018 to present 2025. Thus, translating publications through any of the translation apps was not preferable, as translations that might not be translated correctly can miss nuances. Inclusion and exclusion criteria Studies were included if they: Evaluated AI applications in dental didactic or clinical training. Addressed the ethical, infrastructural, or pedagogical challenges of AI integration. Were peer-reviewed articles or gray literature (e.g., institutional reports, conference proceedings). The exclusion criteria included non-English studies, editorials, and non-AI-related technologies (e.g., standalone VR without AI components). Screening and data extraction After duplicate removal, 510 records underwent title screening, followed by abstract screening of 116 studies and 394 removals, and finally a full-text review of 57 articles. The electronic and manual searches revealed forty-three studies that met the eligibility criteria (see Fig. 1). Data extraction focused on the following: Study design, population, and AI type. Key outcomes (e.g., accuracy, student engagement). Challenges (e.g., bias, infrastructure gaps). Thematic synthesis categorized findings into four domains: personalization, clinical training, institutional efficiency, and ethical governance. Identification Included Figure 1. PRISMA flowchart of study selection Results Study characteristics The 43 included studies (Table 1 ) included comparative (randomized trials) (n = 12), cross-sectional survey (n = 16), review (n = 8), mixed methods (n = 3), prospective (n = 3), and qualitative (n = 1) studies (Fig. 2 ). Geographically, approximately 58% originated from high-income countries (e.g., the USA and Germany), 33% from unspecified/middle-income regions, and 9% from underrepresented regions (e.g., South Asia and Africa). The AI tools evaluated included generative models (e.g., ChatGPT) (32%), diagnostic algorithms (45%), and various other AI tools (23%). Table 1 Overview of Included Studies: Study (Author, Year) Study Design Objectives Population AI Application Key Findings Challenges Identified Future Directions Limitations (Brondani et al., 2024) Mixed-methods 1. Assess if instructors distinguish ChatGPT vs. student reflections. 2. Compare AI vs. human thematic analysis. 20 reflections (10 students, 10 ChatGPT); 3 instructors; 2 qualitative researchers. ChatGPT for generating reflections and performing thematic analysis. - Instructors identified authorship correctly 85% of the time. - Thematic analyses by AI and researchers were comparable. - Risk of academic integrity breaches. - Overreliance may reduce critical thinking. Integrate AI as a supplementary tool (e.g., starting point for assignments). -Purposefully selected reflections may bias results. - Unknown prior AI use by students. (Fang et al., 2024) Randomized controlled trial Compare chatbot (CB) vs. Blackboard (BB) in clinical implant education. 86 predoctoral dental students (UIC-COD). Custom-developed chatbot for answering clinical queries and protocols. - CB improved timeliness, interaction, and reduced anxiety. - Concerns about incorrect AI information ( P = 0.003). - Accuracy and privacy concerns. - Risk of replacing human interaction. Long-term studies on AI integration; balance technology with human-led education. - Short interaction time (10–15 mins). - Single institution. Chang et al., 2024) Randomized controlled trial Evaluate AI-assisted vs. manual FMS radiograph mounting. 40 third-year dental students (UTSD). AI interface for premounting radiographs; students adjusted AI outputs. - AI reduced time (45.99 s vs. 158.01 s, p < 0.05) but lowered accuracy (45–80% vs. 100%, p < 0.01). - Overreliance on AI observed. - Overreliance reduces diagnostic accuracy. - Potential misdiagnosis risks. Use AI as a postlearning verification tool, not during foundational skill acquisition. - Limited to third-year students at one institution. - Small sample size. (Schropp et al., 2024) Randomized controlled trial Assess the impact of AI software on students' ability to detect enamel-only proximal caries in bitewing radiographs. 74 third-year dental students (Aarhus University). AI software (AssistDent®) for caries detection in bitewing radiographs. - AI training did not significantly improve caries detection. - Overlap of teeth reduced detection accuracy. - Overreliance on AI may reduce diagnostic accuracy. - Ethical concerns with AI use in diagnostics. - Use AI as a supplementary tool for training, not replacement. - Long-term studies on AI's impact on learning outcomes. - Limited to one institution. - Small sample size. - Focused only on enamel caries. (Johnsen and Marchini, 2024) Perspective Explore how AI can facilitate the development of critical thinking outcomes in dental education. Dental educators and students. ChatGPT 3.5 for generating learning outcomes and guidance for critical thinking. - AI provides a rich background on critical thinking but lacks explicit learning guidance. - Context-specific outcomes still require educator input. - AI-generated outcomes may lack depth and specificity. - Risk of overreliance on AI for curriculum design. - Develop AI tools to support educators in creating explicit learning outcomes. - Combine AI with human expertise for curriculum development. - Limited to ChatGPT 3.5. - No empirical data on AI's effectiveness in dental education. (Saghiri, Vakhnovetsky and Nadershahi, 2022) Scoping review Examine AI and immersive digital tools in dental education. Dental educators and students. AI, virtual reality (VR), augmented reality (AR), and haptic technology in dental education. - AI and immersive tools enhance learning in implantology, restorative dentistry, and surgical training. - Limited evidence on AI's role in diagnostics. - Ethical concerns with AI and immersive tools. - Limited research on AI's long-term impact. - Further research on AI's role in diagnostics and personalized learning. - Develop guidelines for AI integration in dental curricula. - Limited to studies up to 2021. - Focused on immersive tools rather than AI alone. (Islam et al., 2022) Review (Framework) Provide a framework for AI adoption in dental education using Bolman and Deal's Four Frames model. Dental educators and administrators. AI integration in dental education using the Four Frames model (Structural, Human Resource, Political, Symbolic). - AI can enhance educational experiences and delivery of care. - Faculty resistance and resource limitations are barriers. - Faculty resistance to AI adoption. - Ethical and infrastructural challenges. - Develop inclusive AI strategic planning committees. - Address ethical and infrastructural barriers. - Focused on AI adoption in dental schools, not specific AI tools. - Limited empirical data. (Ali et al., 2024) Cross-sectional Evaluate ChatGPT's performance on dental assessments. 50 assessment items (MCQs, SAQs, SEQs, true/false, fill in the blanks). ChatGPT (free version) for answering knowledge-based and written assessments. - ChatGPT performed well on MCQs, SAQs, true/false, and fill in the blanks (90–100% accuracy). - SEQs scored lower (70%) due to limited detail. - Reflective reports and research methodology were satisfactory, but critical appraisal of literature was borderline. - ChatGPT cannot process image-based questions. - Limited word count in free version. - Risk of academic dishonesty. - Use ChatGPT as a supplementary tool for learning. - Develop policies to mitigate dishonest use of AI. - Explore paid versions for enhanced capabilities. - Limited to text-based questions. - Small number of assessment items. - Only free version of ChatGPT was used. (Al-Zubaidi et al., 2024) Qualitative Explore faculty readiness for AI integration in dental education. 21 faculty members from dental colleges in South Punjab, Pakistan. AI tools for teaching, learning, and administrative tasks. - Faculty had low AI literacy but saw potential benefits in personalized learning and efficiency. - Concerns included lack of training, infrastructure, and ethical issues. - Lack of awareness and training. - Ethical concerns (data privacy, bias). - Resistance to change. - Provide hands-on workshops and peer-learning sessions. - Develop institutional support and infrastructure. - Incorporate cultural and context-specific AI tools. - Small sample size. - Limited to South Punjab, Pakistan. - Qualitative data may not be generalizable. (Aldowah et al., 2024) Cross-sectional Assess perceptions and knowledge of undergraduate dental students about AI. 165 undergraduate dental students from 20 universities in Saudi Arabia. AI applications in various dental specialties (e.g., periodontics, prosthodontics, oral medicine). − 80.6% of students found AI exciting, but basic knowledge was limited. − 66.6% were aware of AI opportunities and threats. - Senior students had better knowledge of AI applications in specific dental fields. - Limited AI training in dental schools. - Concerns about AI replacing dentists. - Reliance on social media for AI information. - Integrate AI into dental curricula. - Provide more AI training and resources. - Address ethical concerns and biases. - Small sample size. - Limited to Saudi Arabian universities. - Self-reported data may have response bias. (Ayan et al., 2024) Randomized controlled trial Investigate the caries diagnosis performance of dental students after AI training. 40 dental students (second-year) divided into two groups (trained vs. nontrained). Modified YOLOv5 model for detecting proximal caries lesions in bitewing radiographs. - AI-trained students showed significant improvement in accuracy, sensitivity, specificity, and F1 scores ( p < 0.05). - AI training increased labelling time for the trained group. - Overreliance on AI may reduce diagnostic accuracy. - Ethical concerns with AI use. - Use AI as a supplementary tool for training. - Long-term studies on AI's impact. - Small sample size. - Limited to one institution. - Focused on bitewing radiographs. (Bahadir et al., 2024) Comparative Compare diagnostic accuracy of dental students and AI in panoramic radiographs. 50 fourth-year dental students (4DS), 50 final-year dental students (5DS), and an AI application. AI application (DentisToday) for detecting caries, fillings, root canal treatments, etc., in panoramic radiographs. - AI outperformed students in detecting caries, fillings, and extractions ( p 0.05). - AI may replace human interaction. - Ethical concerns with data privacy and bias. - Use AI to confirm and strengthen student diagnoses. - Develop AI tools for clinical training. - Limited to panoramic radiographs. - Single AI application used. - Small sample size. (Li et al., 2025) Cross-sectional Compare ChatGPT-3.5/4 with dental students on periodontal surgery concepts. 134 dental students from West China School of Stomatology. ChatGPT-3.5 and ChatGPT-4 for answering multiple-choice and open-ended questions on periodontal surgery. - ChatGPT-4 required more time than ChatGPT-3.5 but had higher accuracy (20/25 vs. 14/25). - ChatGPT-4's review of student feedback was consistent with teacher's review. - ChatGPT's accuracy was not comparable to students. - Ethical concerns with AI-generated feedback. - Use ChatGPT as a supplementary tool for learning and clinical communication. - Long-term studies on AI's impact on dental education. - Limited to one institution. - Language differences between student and ChatGPT responses. (Busch et al., 2024) Cross-sectional Assess global medical, dental, and veterinary students' attitudes towards AI in education and practice. 4596 students from 192 faculties in 48 countries. Survey on AI knowledge, attitudes, and preferences for AI teaching in healthcare education. - Students had positive attitudes towards AI but limited knowledge. − 76.3% reported no AI courses in their curriculum. - Regional differences in AI education and perceptions. - Lack of AI education in curricula. - Ethical and legal concerns with AI integration. - Integrate AI education into medical curricula. - Address regional differences in AI perceptions and educational needs. - Uneven regional distribution of participants. - Potential selection bias due to online survey design. (Dascalu et al., 2024) comparative Investigate AI-initiated second opinions in advanced caries treatment planning. 3 experienced dentists, 25 dental students, and 290 patients across 6 centers. AI model to predict pulp status following advanced caries treatment. - AI-triggered second opinions improved clinicians' F1-score from 0.586 to 0.645. - AI predictions were hidden to minimize distrust and bias. - Clinicians' distrust of AI predictions. - Risk of erroneous AI outputs. - Develop frameworks for AI-assisted second opinions in clinical practice. - Further research on AI-dentist interaction. - Limited to advanced caries treatment. - Small sample size of clinicians. (Elchaghaby and Wahby, 2025) Cross-sectional Evaluate knowledge, attitudes, and perceptions of Egyptian dental students toward AI. 384 dental students from Cairo University and Egyptian Russian University. Survey on AI knowledge, attitudes, and perceptions. − 49% had basic knowledge of AI principles; 48% were aware of AI usage in dentistry. - Most students agreed on AI's leading role in dentistry but disagreed on AI replacing dentists. - Limited AI training in dental schools. - Concerns about AI replacing human dentists. - Incorporate AI applications into undergraduate and postgraduate dental training. - Enhance AI literacy among dental students. - Limited to two universities in Egypt. - Self-reported data may have response bias. (Elnagar et al., 2024) Narrative review Explore the implications of ChatGPT on dental education, including admissions, learning, and research. Dental educators, students, and researchers. ChatGPT for generating essays, answering queries, and facilitating research. - ChatGPT can assist in dental school admissions but raises concerns about plagiarism. - AI can enhance learning but may hinder critical thinking. - AI holds promise for research but lacks domain-specific knowledge. - Ethical concerns with AI-generated content. - Risk of misinformation and bias. - Develop ethical guidelines for AI use in dental education. - Integrate AI responsibly into curricula. - Focused on ChatGPT; limited to narrative review. - Lack of empirical data. (Fitzek and Choi, 2024) Cross-sectional Investigate perceptions of German-speaking medical and dental students regarding AI in healthcare. 409 medical and dental students from Austria, Germany, and Switzerland. Survey on AI familiarity, attitudes, and training experiences. - Only 18.2% of students received formal AI training. - Positive correlations between tech-savviness and AI familiarity. - Dental students had slightly more positive attitudes toward AI. - Lack of comprehensive AI education in curricula. - Ethical and infrastructural challenges. - Integrate AI education into medical and dental curricula. - Address ethical and infrastructural barriers. - Potential selection bias due to social media recruitment. - Self-reported data may have response bias. (Gowdar et al., 2024) Cross-sectional Assess awareness and attitude toward AI among dental students and practitioners. 100 dental students, 100 practitioners (Alkharj, Saudi Arabia) AI in radiological diagnosis, cancer detection, record maintenance. − 33% aware of AI principles; 68% aware of AI uses in dentistry. − 87% believe AI aids radiological diagnosis. − 56.5% agree AI helps in cancer detection. - Low AI literacy (74% unaware of deep learning models). - Ethical concerns (93% worried about violations). - Infrastructure and training gaps. - Integrate AI training into curricula. - Address ethical and infrastructural barriers. - Small sample size. - Regional focus (Alkharj). - Self-reported data may have response bias. (Guler, Yalcin and Gulsun, 2024) Cross-sectional Evaluate opinions on AI in craniomaxillofacial surgery and dentistry. 296 dental students (Dicle University, Turkiye) AI in diagnostics, treatment planning, and imaging analysis. − 74% believe AI will advance dentistry and surgery. − 87.8% willing to use AI in practice. - No significant knowledge difference between academic years. - Limited AI training in curricula. - Concerns about AI replacing human interaction. - Include AI in dental education. - Develop frameworks for AI-dentist collaboration. - Single-institution data. - Potential selection bias (online survey). - Focus on surgery-specific AI applications. (Hammoudi Halat et al., 2024) Cross-sectional Assess AI readiness, perceptions, and educational needs among dental students. 94 dental students (Qatar University) MAIRS scale for AI readiness; topics: AI in healthcare, radiology, research. - Moderate AI readiness (3.3/5). - High need for AI education (84% prioritized AI in healthcare). - Concerns about ethical risks (e.g., data privacy). - Low readiness in cognition/ability domains. - Ethical and infrastructural challenges. - Curriculum updates to address AI literacy. - Foster ethical AI use and interdisciplinary training. - Single-center study. - Small sample size. - Self-reported data. - Limited generalizability. (Hegde et al., 2024) Cross-sectional Evaluate Australian dentists' and students' attitudes toward AI in dentistry. 155 Australian dentists and dental students General AI tools in clinical dentistry − 54.8% aware of AI applications; 70.3% could not name specific software. - 91.6% viewed AI as supportive; concerns included job displacement and mistrust in accuracy. - Limited AI literacy. - Ethical concerns (data privacy, bias). - Resistance to change. Integrate AI into curricula; foster interdisciplinary collaborations; provide hands-on training. - Small sample size. - Self-reported data bias. - Regional focus (Australia). (Hultgren et al., 2023) Cross-sectional Compare ChatGPT-3.5 vs. teachers in answering questions and creating ILOs. 22 dental students and teachers in Sweden ChatGPT-3.5 for Q&A and ILO generation - ChatGPT answered questions comparably to teachers but with more elaboration. - Generated ILOs were often irrelevant or too advanced. - Risk of incorrect/misleading information. - Overreliance reduces critical thinking. Use AI as a supplementary tool; refine prompts for specificity. - Limited to ChatGPT-3.5. - Short interaction time. - Single institution. (Künzle and Paris, 2024) Comparative Assess LLM performance on restorative dentistry and endodontics assessments. 151 dental exam questions ChatGPT-3.5, 4.0, 4.0o; Gemini 1.0 - ChatGPT-4.0° achieved highest accuracy (72%). - Performance varied by subfield (e.g., direct restorations: 84%). - Overreliance risks. - Ethical concerns (data privacy, bias). - Hallucination in responses. Use AI cautiously for exam preparation; update models with dental-specific data. - Single institution. - Focused on text-based questions. - Small sample size. (Lin, Tan and Hashim, 2024) Cross-sectional Explore ethical perceptions of AI in clinical decision-making. 165 Malaysian dental students General AI algorithms in clinical dentistry - Students had positive perceptions of AI ethics. - Females prioritized patient consent/privacy more than males. - Limited understanding of algorithmic transparency. - Risk of data breaches. Integrate ethics education into curricula; address cultural/contextual biases in AI tools. - Single institution. - Self-reported bias. - Lack of qualitative data. (Ozbey and Yasa, 2025) Cross-sectional Evaluate perceptions/attitudes toward AI based on personality traits. 83 dental students (Turkey) General AI tools for radiograph evaluation - Personality traits (Openness/Agreeableness) influenced positive attitudes. - Males more familiar with AI than females. - Participants saw AI as helpful but distrusted it over dentists. - Ethical concerns (data privacy, bias). - Resistance to change. - Tailor educational strategies to personality traits. - Foster interdisciplinary training. - Small sample size. - Self-reported data bias. - Single-institution focus (Turkey). (Qamar et al., 2024) Cross-sectional Assess ChatGPT adoption in academic practices. 315 medical/dental students (Pakistan) ChatGPT for case-based learning/clarifications − 61% used ChatGPT; 85.7% reported knowledge acquisition. - Perceived usefulness score: 17.93 ± 5.08. - Concerns about incorrect information. - Accuracy/reliability issues. - Academic dishonesty risks. Integrate AI with traditional teaching. - Address ethical/infrastructural barriers. - Self-reported data bias. - Regional focus (Pakistan). - Uneven distribution of participants. (Qutieshat et al., 2024) Mixed method Compare diagnostic accuracy of students vs. AI in endodontics. 109 dental students (Oman) Modified ChatGPT 4 for pulpal/apical diagnoses - ChatGPT accuracy: 99% vs. students (77–79%). - Median accuracy: ChatGPT 100% vs. students 82–85%. - Overreliance may hinder critical thinking. - Ethical/legal concerns. Use AI as a supplementary tool. - Update models with dental-specific data. - No radiographs included. - Single-institution data. - Small sample size. (Ramezanzade et al., 2023) Comparative Compare AI vs. dental students in predicting pulp exposure during caries excavation. 25 dental students, 3 dentists, 290 patients. Multipath neural network (ResNet-50 + clinical data) to predict pulp exposure. - AI outperformed students (F1 = 0.71 vs. 0.61). - Students with AI predictions showed marginal improvement ( p = 0.054). - Black-box AI reduced trust. - Class imbalance (more "no exposure" cases). Develop explainable AI (XAI) for clinical transparency. - Address dataset bias. - Small sample size of clinicians. - Limited to bitewing radiographs. - Ethical concerns with AI reliance. (Rampf et al., 2024) Randomized clinical trial Compare elaborated feedback (eF) vs. knowledge of results feedback (KOR) for radiographic diagnostics, with AI validation. 55 fourth-year dental students (Heidelberg University). AI (dentalXrai Pro 3.0) for caries detection on bitewing radiographs. - eF group performed better than KOR group in detecting enamel caries ( p = 0.037) and apical periodontitis ( p = 0.003). - AI achieved near-perfect accuracy (F1 = 0.976). - Ethical concerns with AI replacing expert assessments. - Limited generalizability of AI to primary teeth/periapical lesions. - Use AI to supplement expert feedback in radiology education. - Validate AI in diverse clinical scenarios. - Small sample size. - AI not validated for primary teeth or periapical lesions. - Single-center study. (Roganovic, 2024) Mix methods Investigate how familiarity with ChatGPT features modifies dental students' expectations and learning outcomes. 104 third-year dental students (University of Belgrade). ChatGPT-3.5 for pharmacology learning (side effects of drugs). - Students using ChatGPT (YG group) scored higher on pharmacology quizzes than those not using it (NN group). - Reading ChatGPT descriptions altered expectations but did not increase trust in AI. - Low student willingness to use ChatGPT. - AI description-induced cognitive bias. - Review AI system descriptions for accuracy before educational use. - Address cognitive bias in AI-human interaction studies. - Small sample size for ChatGPT users (YG group). - Limited to pharmacology topics. - Potential selection bias. (Saravia-Rojas et al., 2024) Cross-sectional Evaluate the influence of ChatGPT on academic tasks performed by undergraduate dental students. 55 fourth-year undergraduate dental students (University of Cayetano Heredia, Lima, Peru). ChatGPT for scientific writing assignments. - Traditional method yielded higher scores than ChatGPT ( p = 0.019). - ChatGPT improved productivity but lacked depth in evidence utilization and argument evaluation. - Risk of plagiarism and reduced originality. - Overreliance on AI may hinder critical thinking. Use ChatGPT as a supplementary tool for academic tasks, not a replacement for traditional methods. - Small sample size. - Limited to one institution. - Focused on scientific writing tasks. (Schoenhof et al., 2024) Randomized controlled trial Investigate the use of synthetic panoramic radiographs (syPRs) created by GANs in dental education. 54 medical professionals and 33 dentistry students. Synthetic PRs (syPRs) generated using StyleGAN2-ADA for teaching and research. - syPRs were indistinguishable from real PRs in 78.2% of cases. - Image quality was rated as moderate (median 6/10). − 11 out of 14 items in syPR interpretation showed agreement. - Limited training data (9599 real images). - Logical errors in synthetic images. - Use syPRs for teaching and research without relying on patient-related data. - Upscale training datasets for AI-based diagnostic systems. - Limited to one institution. - Small sample size. - Focused on panoramic radiographs. (Schwendicke et al., 2023) Review Define a core curriculum for AI in oral and dental healthcare education. Dental educators, students, and researchers. AI integration in dental education using a structured curriculum. - Four domains of learning outcomes: AI basics, use cases, evaluation, and governance. - Most outcomes were on the "knowledge" level. - Limited AI literacy among dental professionals. - Ethical and infrastructural challenges. - Integrate AI education into dental curricula. - Foster interdisciplinary collaborations. - Limited to theoretical framework. - Lack of empirical data on curriculum implementation. (Stephan et al., 2024) Comparative Assess effectiveness of ChatGPT in generating radiology reports from dental panoramic radiographs. 100 dental students (University Medical Centre Mainz, Germany) ChatGPT for generating radiology reports based on diagnostic findings. - AI-generated reports showed high textual similarity to reference reports but lacked critical diagnostic information. - AI reports were error-free and matched readability of student-generated reports. - Challenges in accuracy and reliability of AI-generated reports. - Need for prompt refinement. Refine AI algorithms and prompt design to optimize medical reporting. - Limited to panoramic radiographs. - Small sample size. (Tadinada et al., 2023) Perspective Propose a road map for integrating AI, VR, and AR into dental curricula. Dental educators and students AI, VR, AR, and MR for creating agile dental curricula. - AI and immersive tools enhance learning in implantology, restorative dentistry, and surgical training. - Limited evidence on AI's role in diagnostics. - Ethical concerns with AI and immersive tools. - Limited research on AI's long-term impact. Further research on AI's role in diagnostics and personalized learning. - Limited to theoretical framework. - Lack of empirical data on curriculum implementation. (Mahrous et al., 2023) Comparative Compare student performance in RPD design with and without AiDental software. 73 second-year dental students (University of Iowa) AiDental software for RPD design and game-based learning. - AiDental group achieved higher grades (A or B) compared to the conventional group. - Students perceived the software as beneficial for practice, understanding, and feedback. - Lack of detailed explanations for incorrect answers. - Graphics not appealing. - Use AI and gamification techniques in dental education. - Future research on grading module. - Small sample size. - Limited to one institution. - Focus on RPD design. (Thurzo et al., 2023) Review Provide an overview of AI's impact on dental education and propose curriculum updates. Dental educators, students, researchers. ChatGPT, Midjourney for educational tasks. - AI literacy among educators is limited. - Generative AI (e.g., ChatGPT) disrupts traditional education methods. - Ethical and infrastructural barriers hinder adoption. - Faculty readiness. - Ethical/legal implications of AI-generated content. - Technological overhang. - Integrate AI education into curricula. - Develop guidelines for responsible AI use. - Rapidly evolving field. - Lack of empirical data on curriculum implementation. (Yilmaz, Erdem and Uygun, 2024) Cross-sectional Assess AI knowledge, attitudes, and application perspectives among dental students. 335 undergraduate and 62 specialty students (Turkey). General AI tools in clinical dentistry. - Students recognize AI’s advantages in data integration and diagnostics. - Concerns about reduced patient empathy and accountability for errors. - Strong support for AI courses in curricula. - Academic integrity risks. - Overreliance on AI. - Liability for machine errors. - Balance AI with human expertise. - Prioritize patient privacy and data security. - Regional focus (Turkey). - Self-reported data bias. - Small sample size for specialty students. (Amiri et al., 2024) Systematic review & meta-analysis Investigate knowledge and attitudes of medical, dental, and nursing students toward AI. 8491 participants from 22 studies (medical, dental, nursing students). General AI tools in healthcare educat ion. - Pooled knowledge proportion: 0.44 (44% moderate knowledge). - Pooled attitude proportion: 0.65 (65% positive attitudes). - Regional disparities in AI literacy (developed > developing nations). - Knowledge gaps in AI principles. - Ethical concerns (data privacy, job displacement). - Resistance to AI adoption in developing regions. - Integrate AI education into curricula. - Address ethical concerns (bias, transparency). - Conduct longitudinal studies on AI's impact postgraduation. - Language bias (English-only studies). - Heterogeneous methodologies across included studies. - No skill-based data for meta-analysis. (Annamma et al., 2024) Scoping review Identify challenges in dental education systems globally. 12 selected studies (2019–2024) addressing dental education challenges. AI/VR/AR integration in dental curricula. - Institutional challenges: outdated infrastructure, faculty shortages, poor curriculum alignment. - Student challenges: inadequate clinical exposure, stress, lack of competency in geriatric/specialized care. - Technology gaps: limited training in AI/digital tools. - Outdated curricula failing to address modern dental practice needs. - Faculty resistance to AI integration. - Employment struggles for graduates in oversaturated markets. - Curriculum reform to include AI/digital dentistry. - Expand community-based training. - Foster faculty development programs. - Implement competency-based education models. - Limited to English-language studies. - Qualitative focus; lacks quantitative analysis. - Narrow timeframe (2019–2024). (Kim et al., 2023) Perspective Discuss ethical integration of AI into dental curricula. Dental educators, students AI in image analysis, EHR, clinical decision-making. - Emphasizes need for AI literacy, ethical frameworks, and critical thinking. - Curriculum must address biases and limitations. - Ethical concerns, data privacy, academic integrity. - Risk of overreliance on AI. Incremental AI integration, multidisciplinary collaboration, policy updates. Theoretical focus; lacks empirical data; limited tool-specific guidance. (Uribe, Maldupa and Schwendicke, 2025) Scoping Review Summarize guidelines for GenAI in dental education. 31 documents from 15 countries ChatGPT, DALL-E for teaching, assessments, administrative tasks. - Ethical use, transparency, and academic integrity prioritized. - No dental-specific guidelines identified. - Academic misconduct, AI hallucinations, data privacy. - Inequitable access to AI resources. Develop dental-specific guidelines, promote AI literacy, invest in secure platforms. - Selection bias in document inclusion. - limited empirical validation of recommendations. Table 1 : Overview of included studies [See additional file 1] Risk of bias (quality assessment): The relevance and eligibility of the articles included in this scoping review were assessed and critically appraised by one of the researchers, and AK was confirmed by a second researcher, RA. Thus, the risk of bias was eliminated (Appendix 2). The results revealed that 75% of the studies presented a low-to-moderate risk of bias, and 25% presented a high risk. The critical appraisal of the included studies involved the use of multiple tools. For example, for randomized trials, the Cochrane RoB 2.0 [ 48 ]; for cross-sectional surveys, the JBI Cross-Sectional Checklist [ 49 ]; for mixed methods, the Mixed Methods Appraisal Tool (MMAT) [ 50 ]; for experimental/quasiexperimental methods, the JBI Experimental Checklist [ 51 ]; and for reviews/Commentaries, the SANRA [ 52 ]. Key observations of the critical appraisal: Twenty-five percent of studies exhibited a high risk of bias, primarily stemming from subjective outcomes, narrow scopes (e.g., single AI applications such as ChatGPT), or insufficient validation of AI tools. For example, experimental studies testing novel AI interfaces often lack longitudinal follow-up or clinical validation, limiting their translational relevance. Additionally, a recurring weakness across designs was the geographic imbalance, with 58% of studies originating from high-income countries, raising concerns about the applicability of findings to low-resource settings where AI adoption may face infrastructural and ethical barriers. A total of 45% of the studies demonstrated moderate risk of bias, often due to reliance on self-reported data or single-centre designs that limit generalizability. For example, cross-sectional surveys exploring faculty or student perceptions of AI, while valuable for capturing attitudes, frequently employ convenience sampling or lack control groups, introducing potential confounding factors. Thirty percent of the studies were rated as having a low risk of bias, characterized by rigorous methodologies such as preregistered protocols, blinded outcome assessments, or validated measurement tools. Examples include randomized trials comparing AI-generated feedback to traditional methods (e.g., Rampf et al., 2024) and multicenter diagnostic accuracy studies using standardized reference standards (e.g., Schropp et al., 2024). Finally, the appraisal highlighted critical gaps in ethical transparency and data diversity. Few studies have addressed algorithmic bias mitigation or disclosed how training datasets were curated, echoing broader concerns in AI research about reproducibility and equity. Strengths included interdisciplinary adoption or collaboration in mixed methods designs (e.g., Uribe et al., 2025), which enriched contextual insights into AI’s role in curriculum development. Key findings and AI trends in dental education: 1. AI-driven personalization and adaptive learning Generative AI (e.g., ChatGPT) reduces grading time by 45% and improves reflective learning outcomes and dynamic content adjustment on the basis of student performance [3, 53]. However, 33% of studies reported algorithmic bias in feedback systems due to nonrepresentative training data [46] in addition to ethical frameworks and curriculum design challenges [47]. The success metrics of these studies revealed improved student engagement (85% accuracy in thematic analysis by AI vs. human instructors) [3]. 2. AI in clinical and radiographic training AI tools achieved 99% accuracy in caries detection versus 77–79% accuracy for students [35]. The success metrics of AI-trained students revealed significant improvements in labelling time ( p < 0.05) [3]. However, models trained on homogeneous datasets (e.g., European populations) underperformed in diverse cohorts [21], resulting in limited generalizability of institution-specific datasets and overreliance risks. 3. Institutional efficiency and curriculum design Automated scheduling reduced administrative workloads by 30%, but only 18% of institutions had updated curricula to include AI literacy modules [46]. 4. Ethical governance and regulatory compliance Data privacy breaches were reported to occur in 24% of studies, and 41% reported faculty resistance to AI adoption [32]. In addition, there is a lack of dental-specific guidelines and a need for policy development [3]. Discussion Synthesis of findings This scoping review identifies the transformative potential of AI in dental education while underscoring critical challenges. Full-text PDF articles were downloaded, and NIVIVO qualitative data analysis software was used. Four major themes emerged, aligning with prior research but revealing nuanced gaps: 1. Theme 1: Balancing efficiency and critical thinking While generative AI enhances efficiency, overreliance risk diminishes students’ analytical skills. For example, Chang et al. (2024) reported that students using ChatGPT for case studies prioritized speed over diagnostic depth, mirroring concerns in medical education [54]. This suggests a need for hybrid pedagogies that pair AI tools with guided critical reflection, a strategy successfully implemented in problem-based learning (PBL) models at Malmö University in 1990 [55]. 2. Theme 2: The generalizability challenge The diagnostic accuracy of AI surpasses student performance in controlled settings, but its real-world applicability is limited by dataset homogeneity. For example, AI trained on European radiographs faltered in detecting pathologies in Southeast Asian populations [21]. This echoes challenges in oncology, where biased algorithms misdiagnose underrepresented groups [56]. To address this, institutions may adopt federated learning frameworks to train models on diverse, decentralized datasets [57]. 3. Theme 3: The curriculum gap Despite AI’s administrative benefits, curricular integration lags. Only 18% of dental schools have incorporated AI literacy, whereas 42% of medical schools do [46]. This disparity may stem from the limited AI proficiency of faculty, a barrier also observed in nursing education [58]. Partnering with tech firms for faculty training, as done by the Harvard School of Dental Medicine [59], could accelerate competency development. 4. Theme 4: From principles to practice While frameworks such as the ANSI/ADA Standard No. 1110-1:2025 [60] promote algorithmic transparency, 63% of studies lacked independent audits. This gap mirrors regulatory shortcomings in AI-driven radiology, where opaque algorithms have led to misdiagnoses [1]. Implementing mandatory third-party validation, akin to the EU’s GDPR compliance checks, could mitigate risks [61]. Accreditation standards and legislation for AI integration in dental education The rapid integration of AI into dental education demands robust accreditation frameworks and legislation to ensure ethical, equitable, and effective adoption. Current accreditation bodies, such as the Commission on Dental Accreditation (CODA), the General Dental Council (GDC), the Association for Dental Education in Europe (ADEE), and the National Center for Academic Accreditation & Evaluation (NCAAA), must evolve to address AI-specific competencies, including data literacy, algorithmic transparency, and ethical decision-making. For example, the ANSI/ADA Standard No. 1110-1:2025 provides a template for standardizing AI validation in radiographic analysis, emphasizing third-party datasets to mitigate bias and ensure generalizability [60]. Similarly, the proposed ISO 18374, the first international AI standard for dentistry, prioritizes transparency in algorithm design and patient population specificity, which could be adapted to educational contexts to govern AI tools used in simulations or diagnostic training [60]. These standards highlight the necessity of aligning accreditation criteria with technical rigor and ethical accountability, ensuring that AI tools meet both clinical and pedagogical benchmarks. Legislative efforts, such as the EU AI Act , offer a blueprint for balancing innovation with safeguards. By classifying AI systems into risk tiers and mandating human oversight for high-risk applications (e.g., diagnostic tools), the Act ensures that AI remains a complementary tool rather than a replacement for clinical judgment [62]. Dental education could adopt similar risk stratification, requiring faculty to validate AI-generated assessments or treatment plans against human expertise. Furthermore, the PULSE framework [63] , designed for ethical AI-augmented clinical decision support, underscores the importance of patient consent and data contextualization. Translating this into education, institutions could mandate that AI curricula include modules on informed consent for data usage in training algorithms, mirroring real-world ethical practices. Recommendations for AI adoption in dental education 1. Update accreditation standards to include AI competencies - CODA, GDC, ADEE, NCAAA and similar bodies should mandate AI literacy as a core competency, requiring students to demonstrate proficiency in interpreting AI outputs, identifying algorithmic bias, and applying ethical principles, e.g., transparency and justice [62, 64, 65]. For example, the Montreal Declaration’s ten ethical principles for AI, such as nonmaleficence and privacy, could be integrated into accreditation criteria to ensure that graduates uphold these values in practice. - Incorporate ADA Technical Report No. 1109:2025’s emphasis on third-party validation into curriculum design, requiring institutions to use externally validated AI tools for assessments or simulations [66]. 2. Legislate interdisciplinary collaboration and continuous monitoring - The establishment of legislation requires dental schools to collaborate with data scientists, ethicists, and regulatory bodies during AI tool development. The EU AI Act’s requirement for human oversight in high-risk applications could be adapted to educational settings, ensuring that faculty review AI-generated feedback before it impacts student evaluations [62]. - Continuous monitoring mechanisms, such as the PULSE framework’s iterative bias surveillance, should be implemented to audit AI tools for demographic disparities in performance, e.g., accuracy variations across ethnic groups [63]. 3. Development of hybrid assessment models - Combine traditional assessments with AI-driven evaluations to preserve critical thinking. For example, AI is used to grade radiographic interpretations but requires students to defend their diagnoses via oral exams, mirroring the problem-based learning (PBL) model from Malmö University, which balances technology with human mentorship[64]. 4. Creating global ethical guidelines for educational AI - Leveraging the WHO’s Global Strategy on Human Resources for Health to harmonize AI ethics in dental education globally ensures consistency in data privacy and equity standards [67]. This could include universal protocols for student data anonymization and restrictions on commercial AI tools lacking transparency. 5. Investment in faculty training and infrastructure - Institutions should be mandated to allocate funding for AI workshops and partnerships with tech developers, as seen in the ADA Standards Program , which engages clinicians, academics, and industry experts in guideline development [66]. - Cloud-based platforms compliant with GDPR and the American Health Insurance Portability and Accountability Act ( HIPAA) can be adopted for secure data storage, enabling scalable AI integration without compromising student or patient privacy [63]. Study limitations · Temporal bias: All included studies were published in or post-2022, limiting insights into long-term AI impacts. · Geographic imbalance: Underrepresentation of low-resource settings (9% of studies) that may restrict generalizability. Conclusion There is a dearth of evidence that dental educators and practitioners genuinely use AI tools to communicate sensitive patient and student data. However, there is a lack of academic journals that explore or even acknowledge the digital security considerations required to ensure the needed confidentiality of the data. The integration of AI into dental education offers unparalleled opportunities to enhance learning outcomes, streamline institutional workflows, and prepare students for technologically advanced clinical environments. However, this transformation requires a deliberate focus on ethical governance and regulatory rigor. Current accreditation standards must evolve to include AI-specific competencies, ensuring that graduates can critically evaluate AI tools and address algorithmic biases. Legislative frameworks should mandate transparency in AI design and human oversight in high-risk applications, such as diagnostics and assessments. Successful adoption hinges on interdisciplinary collaboration that integrates insights from data science, ethics, and pedagogy and investment in faculty training and secure infrastructure. Hybrid models, which combine AI-driven efficiency with traditional mentorship, can preserve critical thinking while leveraging AI’s analytical strengths. Ultimately, AI’s role in dental education must be complementary, not substitutive. By embedding accountability into accreditation, fostering equity through legislation, and prioritizing continuous monitoring, the dental community can ensure that AI serves as a catalyst for innovation while upholding the profession’s ethical foundations. Future research should focus on longitudinal studies to evaluate AI’s long-term impact on clinical competence and patient outcomes, ensuring that this technological revolution remains anchored in evidence and equity. Declarations Ethical approval Since this is a scoping literature review, no data were collected from humans. An ethical approval process was not needed. Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request. Competing interests The authors declare that they have no competing interests. Funding There are no financial conflicts of interest to disclose. Authors’ contributions AK and RA both contributed to the synthesis, analysis, review, and writing of the evidence. Both authors read and approved the final manuscript. Clinical trial number: not applicable. References Topol EJ. Deep medicine: how artificial intelligence can make healthcare human again, First edition. edn. New York: Basic Books; 2019. Zawacki-Richter O, Marín VI, Bond M, Gouverneur F. Systematic review of research on artificial intelligence applications in higher education – where are the educators? Int J Educational Technol High Educ. 2019;16(1):1–27. Uribe SE, Maldupa I, Schwendicke F. Integrating Generative AI in Dental Education: A Scoping Review of Current Practices and Recommendations. Eur J Dent Educ 2025. Schwendicke F, Chaurasia A, Wiegand T, Uribe SE, Fontana M, Akota I, Tryfonos O, Krois J. Artificial intelligence for oral and dental healthcare: Core education curriculum. J Dent. 2023;128:104363–104363. Saghiri MA, Vakhnovetsky J, Nadershahi N. Scoping review of artificial intelligence and immersive digital tools in dental education. J Dent Educ. 2022;86(6):736–50. Carter S, Win K, Wang L, Rogers W, Richards B, Houssami N. 65 Ethical, legal and social implications of artificial intelligence systems for screening and diagnosis. BMJ evidence-based Med. 2019;24(Suppl 2):A37. Thurzo A, Strunga M, Urban R, Surovková J, Afrashtehfar KI. Impact of Artificial Intelligence on Dental Education: A Review and Guide for Curriculum Update. Educ Sci. 2023;13(2):150. Tricco AC, Lillie E, Zarin W, O'Brien KK, Colquhoun H, Levac D, Moher D, Peters MDJ, Horsley T, Weeks L, et al. PRISMA extension for scoping reviews (PRISMA-ScR): Checklist and explanation. Ann Intern Med. 2018;169(7):467–73. Brondani M, Alves C, Ribeiro C, Braga MM, Garcia RCM, Ardenghi T, Pattanaporn K. Artificial intelligence, ChatGPT, and dental education: Implications for reflective assignments and qualitative research. J Dent Educ. 2024;88(12):1671–80. Fang Q, Reynaldi R, Araminta AS, Kamal I, Saini P, Afshari FS, Tan S-C, Yuan JC-C, Qomariyah NN, Sukotjo C. Artificial Intelligence (AI)-driven dental education: Exploring the role of chatbots in a clinical learning environment. J Prosthet Dent 2024. Chang J, Bliss L, Angelov N, Glick A. Artificial intelligence-assisted full‐mouth radiograph mounting in dental education. J Dent Educ. 2024;88(7):933–9. Schropp L, Sørensen APS, Devlin H, Matzen LH. Use of artificial intelligence software in dental education: A study on assisted proximal caries assessment in bitewing radiographs. Eur J Dent Educ. 2024;28(2):490–6. Johnsen DC, Marchini L. Artificial intelligence to develop outcomes for critical thinking: A helping start and still up to the educator to develop the final outcome. Eur J Dent Educ. 2024;28(4):877–9. Islam NM, Laughter L, Sadid-Zadeh R, Smith C, Dolan TA, Crain G, Squarize CH. Adopting artificial intelligence in dental education: A model for academic leadership and innovation. J Dent Educ. 2022;86(11):1545–51. Ali K, Barhom N, Tamimi F, Duggal M. ChatGPT—A double-edged sword for healthcare education? Implications for assessments of dental students. Eur J Dent Educ. 2024;28(1):206–11. Al-Zubaidi SM, Shaikh GM, Malik A, Abideen MZU, Tareen J, Alzahrani NSA, Siddiqui AA. Exploring Faculty Preparedness for Artificial Intelligence-Driven Dental Education: A Multicentre Study. Curēus (Palo Alto CA). 2024;16(7):e64377. Aldowah O, Almakrami A, Alghuwaynim Y, Alhutaylah M, Almansour A, Alswedan A, Alshahrani F, Alqarni S, Alkasi Y. Perceptions and Knowledge of Undergraduate Dental Students about Artificial Intelligence in Dental Schools: A Cross-sectional Study. J Contemp Dent Pract. 2024;25(2):149–55. Ayan E, Bayraktar Y, Çelik Ç, Ayhan B. Dental student application of artificial intelligence technology in detecting proximal caries lesions. J Dent Educ. 2024;88(4):490–500. Bahadir HS, Keskin NB, Cakmak ESK, Gunec G, Cesur Aydin K, Sari SK. Comparison of diagnoses made by dentistry students and by artificial intelligence dentists. J Dent Educ 2024. Li C, Zhang J, Abdul-Masih J, Zhang S, Yang J. Performance of ChatGPT and Dental Students on Concepts of Periodontal Surgery. Eur J Dent Educ. 2025;29(1):36–43. Busch F, Hoffmann L, Truhn D, Ortiz-Prado E, Makowski MR, Bressem KK, Adams LC. Global cross-sectional student survey on AI in medical, dental, and veterinary education and practice at 192 faculties. BMC Med Educ. 2024;24(1):1066–1020. Dascalu T, Ramezanzade S, Bakhshandeh A, Bjorndal L, Ibragimov B. AI-initiated second opinions: a framework for advanced caries treatment planning. BMC Oral Health. 2024;24(1):772–710. Elchaghaby M, Wahby R. Knowledge, attitudes, and perceptions of a group of Egyptian dental students toward artificial intelligence: a cross-sectional study. BMC Oral Health. 2025;25(1):11–7. Elnagar MH, Yadav S, Venugopalan SR, Lee MK, Oubaidin M, Rampa S, Allareddy V. ChatGPT and dental education: Opportunities and challenges. Semin Orthod. 2024;30(4):401–4. Fitzek S, Choi K-EA. Shaping future practices: German-speaking medical and dental students' perceptions of artificial intelligence in healthcare. BMC Med Educ. 2024;24(1):844–8. Gowdar IM, Alateeq AA, Alnawfal AMA, Alharbi AFA, Alhabshan AMS, Aldawsari SMS, AlHarbi NAH. Artificial Intelligence and its Awareness and Utilization among Dental Students and Private Dental Practitioners at Alkharj, Saudi Arabia. J Pharm bioallied Sci. 2024;16(Suppl 3):S2264–7. Guler R, Yalcin E, Gulsun B. Evaluation of Attitudes and Perceptions in Students About the Use of Artificial Intelligence in Craniomaxillofacial Surgery. J Craniofac Surg 2024. Hammoudi Halat D, Shami R, Daud A, Sami W, Soltani A, Malki A. Artificial Intelligence Readiness, Perceptions, and Educational Needs Among Dental Students: A Cross-Sectional Study. Clin experimental Dent Res. 2024;10(4):e925. -n/a . Hegde S, Nanayakkara S, Jordan A, Jeha O, Patel U, Luu V, Gao J. Attitudes and Perceptions of Australian Dentists and Dental Students Towards Applications of Artificial Intelligence in Dentistry: A Survey. Eur J Dent Educ 2024. Hultgren C, Lindkvist A, Özenci V, Curbo S. ChatGPT (GPT-3.5) as an assistant tool in microbial pathogenesis studies in Sweden: a cross-sectional comparative study. J educational evaluation health professions. 2023;20:32–32. Künzle P, Paris S. Performance of large language artificial intelligence models on solving restorative dentistry and endodontics student assessments. Clin Oral Invest. 2024;28(11):575. Lin GSS, Tan WW, Hashim H. Students' perceptions towards the ethical considerations of using artificial intelligence algorithms in clinical decision-making. Br Dent J 2024. Ozbey F, Yasa Y. The relationships of personality traits on perceptions and attitudes of dentistry students towards AI. BMC Med Educ. 2025;25(1):26–8. Qamar K, Butt SW, Abaid M, Rashid A, Tahira R, Iqbal AS. Exploring the adoption and utilisation of ChatGPT in everyday academic practices among medical and dental students of Rawalpindi and Islamabad: a study on the impact and perceived effectiveness. J Pak Med Assoc. 2024;74(12):2101–6. Qutieshat A, Al Rusheidi A, Al Ghammari S, Alarabi A, Salem A, Zelihic M. Comparative analysis of diagnostic accuracy in endodontic assessments: dental students vs. artificial intelligence. Diagnosis 2024(0).. Ramezanzade S, Dascalu TL, Ibragimov B, Bakhshandeh A, Bjørndal L. Prediction of pulp exposure before caries excavation using artificial intelligence: Deep learning-based image data versus standard dental radiographs. J Dent. 2023;138:104732–104732. Rampf S, Gehrig H, Möltner A, Fischer MR, Schwendicke F, Huth KC. Radiographical diagnostic competences of dental students using various feedback methods and integrating an artificial intelligence application—A randomized clinical trial. Eur J Dent Educ. 2024;28(4):925–37. Roganovic J. Familiarity with ChatGPT Features Modifies Expectations and Learning Outcomes of Dental Students. Int Dent J. 2024;74(6):1456–62. Saravia-Rojas MÁ, Camarena‐Fonseca AR, León‐Manco R, Geng‐Vivanco R. Artificial intelligence: ChatGPT as a disruptive didactic strategy in dental education. J Dent Educ. 2024;88(6):872–6. Schoenhof R, Schoenhof R, Blumenstock G, Lethaus B, Hoefert S. Synthetic, non-person related panoramic radiographs created by generative adversarial networks in research, clinical, and teaching applications. J Dent. 2024;146:105042. Stephan D, Bertsch A, Burwinkel M, Vinayahalingam S, Al-Nawas B, Kaemmerer P, Thiem DGE. AI in Dental Radiology-Improving the Efficiency of Reporting With ChatGPT: Comparative Study. J Med Internet Res. 2024;26(7):e60684. Tadinada A, Gul G, Godwin L, Al Sakka Y, Crain G, Stanford CM, Johnson J. Utilizing an organizational development framework as a road map for creating a technology-driven agile curriculum in predoctoral dental education. J Dent Educ. 2023;87(3):394–400. Mahrous A, Botsko DL, Elgreatly A, Tsujimoto A, Qian F, Schneider GB. The use of artificial intelligence and game-based learning in removable partial denture design: A comparative study. J Dent Educ. 2023;87(8):1188–99. Yilmaz C, Erdem RZ, Uygun LA. Artificial intelligence knowledge, attitudes and application perspectives of undergraduate and specialty students of faculty of dentistry in Turkey: an online survey research. BMC Med Educ. 2024;24(1):1149–1112. Amiri H, Peiravi S, rezazadeh shojaee Ss, Rouhparvarzamin M, Nateghi MN, Etemadi MH, ShojaeiBaghini M, Musaie F, Anvari MH, Asadi Anar M. Medical, dental, and nursing students’ attitudes and knowledge towards artificial intelligence: a systematic review and meta-analysis. BMC Med Educ. 2024;24(1):412–412. Annamma LM, Varma SR, Abuttayem H, Prasad P, Azim SA, Odah R, George BT, Nair C, Karobari MI. Current challenges in dental education- a scoping review. BMC Med Educ. 2024;24(1):1523–1510. Kim CS, Samaniego CS, Sousa Melo SL, Brachvogel WA, Baskaran K, Rulli D. Artificial intelligence (A.I.) in dental curricula: Ethics and responsible integration. J Dent Educ. 2023;87(11):1570–3. Sterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, Cates CJ, Cheng H-Y, Corbett MS, Eldridge SM, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898. Critical Appraisal Checklist for Analytical Cross Sectional Studies. [ https://jbi.global/sites/default/files/2020-08/Checklist_for_Analytical_Cross_Sectional_Studies.pdf] Hong QN, Fàbregues S, Bartlett G, Boardman F, Cargo M, Dagenais P, Gagnon M-P, Griffiths F, Nicolau B, O’Cathain A, et al. The Mixed Methods Appraisal Tool (MMAT) version 2018 for information professionals and researchers. Educ Inform. 2018;34(4):285–91. Critical Appraisal Checklist for Quasi-Experimental Studies. [ https://jbi.global/sites/default/files/2020-08/Checklist_for_Quasi-Experimental_Appraisal_Tool.pdf] Baethge C, Goldbeck-Wood S, Mertens S. SANRA—a scale for the quality assessment of narrative review articles. Res Integr Peer Rev. 2019;4(1):5. John RM. AI ethics in dental practice. Br Dent J. 2024;236(9):698–698. Wartman SA, Combs CD. Reimagining medical education in the age of AI. AMA J ethics. 2019;21(2):146–52. Rohlin M, Petersson K, Svensäter G. The Malmö model: a problem-based learning curriculum in undergraduate dental education. Eur J Dent Educ. 1998;2(3):103–14. Obermeyer Z, Powers B, Vogeli C, Mullainathan S. Dissecting racial bias in an algorithm used to manage the health of populations. Science. 2019;366(6464):447–53. Rieke N, Hancox J, Li W, Milletari F, Roth HR, Albarqouni S, Bakas S, Galtier MN, Landman BA, Maier-Hein K. The future of digital health with federated learning. NPJ Digit Med. 2020;3(1):1–7. Buchanan C, Howitt ML, Wilson R, Booth RG, Risling T, Bamford M. Predicted influences of artificial intelligence on nursing education: Scoping review. JMIR Nurs. 2021;4(1):e23933. Exploring how AI. can enhance dental education [ https://www.hsdm.harvard.edu/news/exploring-how-ai-can-enhance-dental-education] Dentistry. Validation Dataset Guidance for Image Analysis Systems Using Artificial Intelligence [ https://www.ada.org/-/media/Project/ADA%20Organization/ADA/ADA-org/Files/Resources/Practice/Dental%20Standards/ADA_1110-1_2025] GDPR compliance checklist. [ https://www.onetrust.com/solutions/gdpr-compliance/] Ducret M, Wahal E, Gruson D, Amrani S, Richert R, Mouncif-Moungache M, Schwendicke F. Trustworthy Artificial Intelligence in Dentistry: Learnings from the EU AI Act. J Dent Res. 2024;103(11):1051–6. Dykstra S, MacDonald M, Beaudry R, Labib D, King M, Feng Y, Flewitt J, Bakal J, Lee B, Dean S, et al. An institutional framework to support ethical fair and equitable artificial intelligence augmented care. npj Digit Med. 2025;8(1):84. Harte M, Carey B, Feng QJ, Alqarni A, Albuquerque R. Transforming undergraduate dental education: the impact of artificial intelligence. Br Dent J. 2025;238(1):57–60. Jobin A, Ienca M, Vayena E. The global landscape of AI ethics guidelines. Nat Mach Intell. 2019;1(9):389–99. ADA technical report calls for external validation. of AI algorithms in dentistry [ https://adanews.ada.org/ada-news/2024/october/ada-technical-report-calls-for-external-validation-of-ai-algorithms-in-dentistry/] Palaniappan K, Lin EYT, Vogel S. Global Regulatory Frameworks for the Use of Artificial Intelligence (AI) in the Healthcare Services Sector. Healthc (Basel) 2024, 12(5). Additional Declarations No competing interests reported. Supplementary Files Additionalfile2.docx Appendix 1: PRISMA ScR Checklist: [See additional file 2] Additionalfile3.docx Appendix 2: Critical appraisal of included studies: [See additional file 3] Cite Share Download PDF Status: Posted Version 1 posted 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-6498214","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":448353190,"identity":"4145db77-3b27-4288-a6d7-a01ea3afc2f3","order_by":0,"name":"Ayman M. Khalifah","email":"data:image/png;base64,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","orcid":"","institution":"University of Manchester","correspondingAuthor":true,"prefix":"","firstName":"Ayman","middleName":"M.","lastName":"Khalifah","suffix":""},{"id":448353191,"identity":"c0638697-5a9e-4204-95bd-2f04bd099367","order_by":1,"name":"Rasha Alafaleg","email":"","orcid":"","institution":"Qassim University","correspondingAuthor":false,"prefix":"","firstName":"Rasha","middleName":"","lastName":"Alafaleg","suffix":""}],"badges":[],"createdAt":"2025-04-21 18:38:09","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6498214/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6498214/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":81953758,"identity":"1e42d225-5f13-46d5-b140-7ca1e72917e6","added_by":"auto","created_at":"2025-05-05 09:40:19","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":26805,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cu\u003e\u003cstrong\u003ePRISMA flowchart of study selection\u003c/strong\u003e\u003c/u\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6498214/v1/c9881590ed6737b76460846b.png"},{"id":81955094,"identity":"899aa747-4f1b-4297-a58e-bc3eb9301e2e","added_by":"auto","created_at":"2025-05-05 09:48:19","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":37617,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cu\u003eDistribution of study designs.\u003c/u\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6498214/v1/804e4241249da6ed344134d2.png"},{"id":84036644,"identity":"7227eff4-92f1-40ce-aa53-2aacd319fd85","added_by":"auto","created_at":"2025-06-06 04:16:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1749966,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6498214/v1/6aaad767-83fe-44aa-811e-d77523af90e6.pdf"},{"id":81956527,"identity":"e6597e54-176e-48d5-a456-e124f96d1d20","added_by":"auto","created_at":"2025-05-05 09:56:19","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":24616,"visible":true,"origin":"","legend":"\u003cp\u003eAppendix 1: PRISMA ScR Checklist: [See additional file 2]\u003c/p\u003e","description":"","filename":"Additionalfile2.docx","url":"https://assets-eu.researchsquare.com/files/rs-6498214/v1/dac18d60fe7caa03907fab5a.docx"},{"id":81956529,"identity":"7be9a7c9-1ea7-4c40-b1c9-2e815fb55f12","added_by":"auto","created_at":"2025-05-05 09:56:19","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":20649,"visible":true,"origin":"","legend":"\u003cp\u003eAppendix 2: Critical appraisal of included studies: [See additional file 3]\u003c/p\u003e","description":"","filename":"Additionalfile3.docx","url":"https://assets-eu.researchsquare.com/files/rs-6498214/v1/61f3b1fcfe390a6e6faf0637.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eArtificial Intelligence in Dental Education: A Scoping Review of Opportunities, Challenges, and Ethical Frameworks for Shaping Accreditation Standards and Future Practice\u003c/p\u003e","fulltext":[{"header":"Background","content":"\u003cp\u003eThe integration of artificial intelligence (AI) into healthcare has revolutionized diagnostic accuracy, patient management, and personalized treatment planning [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. In education, AI-driven tools are reshaping pedagogical approaches by enabling adaptive learning, automated assessments, and immersive simulations [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Dental education, positioned at the intersection of healthcare and pedagogy, stands to benefit significantly from these advancements. However, the rapid adoption of AI also introduces ethical, logistical, and pedagogical challenges that demand systematic exploration[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe current literature highlights AI\u0026rsquo;s potential in dental education, such as virtual reality (VR) simulations for procedural training and generative AI for personalized feedback [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. However, existing studies often focus on isolated applications, with a limited synthesis of broader implications, including faculty readiness, curricular alignment, and ethical governance. For example, while AI-enhanced diagnostics improve efficiency, concerns about algorithmic bias and overreliance on automated systems remain understudied [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Furthermore, disparities in AI adoption between high- and low-resource institutions risk exacerbating global inequities in dental training [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Thus, education and training for health professionals and the need for a policy to be identified for AI implementation are areas of importance for future research.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eThis scoping review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) (Appendix 1) [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. A comprehensive search strategy was employed across PubMed, EMBASE, MEDLINE-Ovid, and Google Scholar, covering studies published between January 2018 and January 2025. The search terms included combinations of \u003cem\u003e\"artificial intelligence,\" \"dental education,\" \"machine learning,\" \"ChatGPT,\"\u003c/em\u003e and \u003cem\u003e\"ethical challenges,\"\u003c/em\u003e \u0026ldquo;with Medical Subject Headings (MeSH) terms applied where applicable. Additionally, manual searches of gray literature, including institutional reports and conference proceedings, were conducted to ensure the broad inclusion of relevant studies. After removing duplicates, 624 records underwent title/abstract screening, followed by a full-text review of 57 articles. Forty-three studies met the eligibility criteria and were included in the final analysis (Fig.\u0026nbsp;1). Data extraction focused on study design, population, AI type, key outcomes, and challenges, with thematic synthesis categorizing findings into four domains: personalization, clinical training, institutional efficiency, and ethical governance.\u003c/p\u003e \u003cp\u003eA scoping review protocol was developed according to the Jonna Briggs Institute framework (JBI) guidelines. Thus, an overview of the literature was mapped to address broader questions. A data extraction (selection and coding) protocol was developed around population/intervention/compare/outcome (PICO).\u003c/p\u003e \u003cp\u003eObjective and research question\u003c/p\u003e \u003cp\u003eThe purpose and objectives of this review are to answer gaps in the literature by systematically mapping AI applications, challenges, and future directions in dental education. The primary research question is as follows: what are the common AI tools used by dental educators at present? \u003cem\u003eHow can AI be integrated into dental curricula to enhance learning outcomes while ensuring ethical, equitable, and pedagogically sound practices?\u003c/em\u003e This study aims to synthesize evidence on AI\u0026rsquo;s educational applications, identify barriers to implementation, and propose strategies for responsible integration aligned with global standards.\u003c/p\u003e \u003cp\u003eIdentifying relevant studies and eligibility criteria\u003c/p\u003e \u003cp\u003eFor this scoping review to be efficient, it will be important to draw upon the full range of published literature from quantitative and qualitative studies to policies and the editorial press. To include from the academic literature to health services and general literature in all ranges of clinical practice to gather evidence that is relevant to dentistry. Publications in the English language were reviewed as most trade publications and gray literature from 2018 to present 2025. Thus, translating publications through any of the translation apps was not preferable, as translations that might not be translated correctly can miss nuances.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eInclusion and exclusion criteria\u003c/h2\u003e \u003cp\u003eStudies were included if they:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eEvaluated AI applications in dental didactic or clinical training.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eAddressed the ethical, infrastructural, or pedagogical challenges of AI integration.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eWere peer-reviewed articles or gray literature (e.g., institutional reports, conference proceedings).\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eThe exclusion criteria included non-English studies, editorials, and non-AI-related technologies (e.g., standalone VR without AI components).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eScreening and data extraction\u003c/h3\u003e\n\u003cp\u003eAfter duplicate removal, 510 records underwent title screening, followed by abstract screening of 116 studies and 394 removals, and finally a full-text review of 57 articles. The electronic and manual searches revealed forty-three studies that met the eligibility criteria (see Fig.\u0026nbsp;1). Data extraction focused on the following:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eStudy design, population, and AI type.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eKey outcomes (e.g., accuracy, student engagement).\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eChallenges (e.g., bias, infrastructure gaps).\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThematic synthesis categorized findings into four domains: personalization, clinical training, institutional efficiency, and ethical governance.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eIdentification\u003c/h3\u003e\n\u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eIncluded\u003c/h3\u003e\n\u003cp\u003e \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eFigure 1. PRISMA flowchart of study selection\u003c/span\u003e \u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eStudy characteristics\u003c/p\u003e\n\u003cp\u003eThe 43 included studies (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e) included comparative (randomized trials) (n\u0026thinsp;=\u0026thinsp;12), cross-sectional survey (n\u0026thinsp;=\u0026thinsp;16), review (n\u0026thinsp;=\u0026thinsp;8), mixed methods (n\u0026thinsp;=\u0026thinsp;3), prospective (n\u0026thinsp;=\u0026thinsp;3), and qualitative (n\u0026thinsp;=\u0026thinsp;1) studies (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). Geographically, approximately 58% originated from high-income countries (e.g., the USA and Germany), 33% from unspecified/middle-income regions, and 9% from underrepresented regions (e.g., South Asia and Africa). The AI tools evaluated included generative models (e.g., ChatGPT) (32%), diagnostic algorithms (45%), and various other AI tools (23%).\u003c/p\u003e\n\u003cp\u003e\u003c/p\u003e\u0026nbsp;\u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eOverview of Included Studies:\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eStudy (Author, Year)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eStudy Design\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eObjectives\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePopulation\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAI Application\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eKey Findings\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eChallenges Identified\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eFuture Directions\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLimitations\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Brondani et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMixed-methods\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1. Assess if instructors distinguish ChatGPT vs. student reflections. 2. Compare AI vs. human thematic analysis.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20 reflections (10 students, 10 ChatGPT); 3 instructors; 2 qualitative researchers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT for generating reflections and performing thematic analysis.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Instructors identified authorship correctly 85% of the time.\u003c/p\u003e\n \u003cp\u003e- Thematic analyses by AI and researchers were comparable.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Risk of academic integrity breaches.\u003c/p\u003e\n \u003cp\u003e- Overreliance may reduce critical thinking.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIntegrate AI as a supplementary tool (e.g., starting point for assignments).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-Purposefully selected reflections may bias results.\u003c/p\u003e\n \u003cp\u003e- Unknown prior AI use by students.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Fang et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRandomized controlled trial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCompare chatbot (CB) vs. Blackboard (BB) in clinical implant education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e86 predoctoral dental students (UIC-COD).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCustom-developed chatbot for answering clinical queries and protocols.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- CB improved timeliness, interaction, and reduced anxiety.\u003c/p\u003e\n \u003cp\u003e- Concerns about incorrect AI information (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Accuracy and privacy concerns.\u003c/p\u003e\n \u003cp\u003e- Risk of replacing human interaction.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLong-term studies on AI integration; balance technology with human-led education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Short interaction time (10\u0026ndash;15 mins).\u003c/p\u003e\n \u003cp\u003e- Single institution.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eChang et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRandomized controlled trial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEvaluate AI-assisted vs. manual FMS radiograph mounting.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e40 third-year dental students (UTSD).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI interface for premounting radiographs; students adjusted AI outputs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI reduced time (45.99 s vs. 158.01 s, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) but lowered accuracy (45\u0026ndash;80% vs. 100%, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01).\u003c/p\u003e\n \u003cp\u003e- Overreliance on AI observed.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Overreliance reduces diagnostic accuracy.\u003c/p\u003e\n \u003cp\u003e- Potential misdiagnosis risks.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eUse AI as a postlearning verification tool, not during foundational skill acquisition.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to third-year students at one institution.\u003c/p\u003e\n \u003cp\u003e- Small sample size.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Schropp et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRandomized controlled trial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAssess the impact of AI software on students\u0026apos; ability to detect enamel-only proximal caries in bitewing radiographs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e74 third-year dental students (Aarhus University).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI software (AssistDent\u0026reg;) for caries detection in bitewing radiographs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI training did not significantly improve caries detection.\u003c/p\u003e\n \u003cp\u003e- Overlap of teeth reduced detection accuracy.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Overreliance on AI may reduce diagnostic accuracy.\u003c/p\u003e\n \u003cp\u003e- Ethical concerns with AI use in diagnostics.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Use AI as a supplementary tool for training, not replacement.\u003c/p\u003e\n \u003cp\u003e- Long-term studies on AI\u0026apos;s impact on learning outcomes.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to one institution.\u003c/p\u003e\n \u003cp\u003e- Small sample size. - Focused only on enamel caries.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Johnsen and Marchini, 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePerspective\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eExplore how AI can facilitate the development of critical thinking outcomes in dental education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDental educators and students.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT 3.5 for generating learning outcomes and guidance for critical thinking.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI provides a rich background on critical thinking but lacks explicit learning guidance.\u003c/p\u003e\n \u003cp\u003e- Context-specific outcomes still require educator input.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI-generated outcomes may lack depth and specificity.\u003c/p\u003e\n \u003cp\u003e- Risk of overreliance on AI for curriculum design.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Develop AI tools to support educators in creating explicit learning outcomes.\u003c/p\u003e\n \u003cp\u003e- Combine AI with human expertise for curriculum development.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to ChatGPT 3.5. - No empirical data on AI\u0026apos;s effectiveness in dental education.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Saghiri, Vakhnovetsky and Nadershahi, 2022)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eScoping review\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eExamine AI and immersive digital tools in dental education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDental educators and students.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI, virtual reality (VR), augmented reality (AR), and haptic technology in dental education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI and immersive tools enhance learning in implantology, restorative dentistry, and surgical training.\u003c/p\u003e\n \u003cp\u003e- Limited evidence on AI\u0026apos;s role in diagnostics.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Ethical concerns with AI and immersive tools.\u003c/p\u003e\n \u003cp\u003e- Limited research on AI\u0026apos;s long-term impact.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Further research on AI\u0026apos;s role in diagnostics and personalized learning.\u003c/p\u003e\n \u003cp\u003e- Develop guidelines for AI integration in dental curricula.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to studies up to 2021.\u003c/p\u003e\n \u003cp\u003e- Focused on immersive tools rather than AI alone.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Islam et al., 2022)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eReview (Framework)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eProvide a framework for AI adoption in dental education using Bolman and Deal\u0026apos;s Four Frames model.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDental educators and administrators.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI integration in dental education using the Four Frames model (Structural, Human Resource, Political, Symbolic).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI can enhance educational experiences and delivery of care.\u003c/p\u003e\n \u003cp\u003e- Faculty resistance and resource limitations are barriers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Faculty resistance to AI adoption.\u003c/p\u003e\n \u003cp\u003e- Ethical and infrastructural challenges.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Develop inclusive AI strategic planning committees.\u003c/p\u003e\n \u003cp\u003e- Address ethical and infrastructural barriers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Focused on AI adoption in dental schools, not specific AI tools.\u003c/p\u003e\n \u003cp\u003e- Limited empirical data.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Ali et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEvaluate ChatGPT\u0026apos;s performance on dental assessments.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e50 assessment items (MCQs, SAQs, SEQs, true/false, fill in the blanks).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT (free version) for answering knowledge-based and written assessments.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- ChatGPT performed well on MCQs, SAQs, true/false, and fill in the blanks (90\u0026ndash;100% accuracy).\u003c/p\u003e\n \u003cp\u003e- SEQs scored lower (70%) due to limited detail.\u003c/p\u003e\n \u003cp\u003e- Reflective reports and research methodology were satisfactory, but critical appraisal of literature was borderline.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- ChatGPT cannot process image-based questions.\u003c/p\u003e\n \u003cp\u003e- Limited word count in free version.\u003c/p\u003e\n \u003cp\u003e- Risk of academic dishonesty.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Use ChatGPT as a supplementary tool for learning.\u003c/p\u003e\n \u003cp\u003e- Develop policies to mitigate dishonest use of AI.\u003c/p\u003e\n \u003cp\u003e- Explore paid versions for enhanced capabilities.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to text-based questions.\u003c/p\u003e\n \u003cp\u003e- Small number of assessment items.\u003c/p\u003e\n \u003cp\u003e- Only free version of ChatGPT was used.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Al-Zubaidi et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQualitative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eExplore faculty readiness for AI integration in dental education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e21 faculty members from dental colleges in South Punjab, Pakistan.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI tools for teaching, learning, and administrative tasks.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Faculty had low AI literacy but saw potential benefits in personalized learning and efficiency.\u003c/p\u003e\n \u003cp\u003e- Concerns included lack of training, infrastructure, and ethical issues.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Lack of awareness and training. - Ethical concerns (data privacy, bias).\u003c/p\u003e\n \u003cp\u003e- Resistance to change.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Provide hands-on workshops and peer-learning sessions.\u003c/p\u003e\n \u003cp\u003e- Develop institutional support and infrastructure. - Incorporate cultural and context-specific AI tools.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size. - Limited to South Punjab, Pakistan.\u003c/p\u003e\n \u003cp\u003e- Qualitative data may not be generalizable.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Aldowah et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAssess perceptions and knowledge of undergraduate dental students about AI.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e165 undergraduate dental students from 20 universities in Saudi Arabia.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI applications in various dental specialties (e.g., periodontics, prosthodontics, oral medicine).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;80.6% of students found AI exciting, but basic knowledge was limited. \u0026minus;\u0026thinsp;66.6% were aware of AI opportunities and threats.\u003c/p\u003e\n \u003cp\u003e- Senior students had better knowledge of AI applications in specific dental fields.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited AI training in dental schools.\u003c/p\u003e\n \u003cp\u003e- Concerns about AI replacing dentists.\u003c/p\u003e\n \u003cp\u003e- Reliance on social media for AI information.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Integrate AI into dental curricula.\u003c/p\u003e\n \u003cp\u003e- Provide more AI training and resources.\u003c/p\u003e\n \u003cp\u003e- Address ethical concerns and biases.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size. - Limited to Saudi Arabian universities. - Self-reported data may have response bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Ayan et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRandomized controlled trial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInvestigate the caries diagnosis performance of dental students after AI training.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e40 dental students (second-year) divided into two groups (trained vs. nontrained).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eModified YOLOv5 model for detecting proximal caries lesions in bitewing radiographs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI-trained students showed significant improvement in accuracy, sensitivity, specificity, and F1 scores (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e\n \u003cp\u003e- AI training increased labelling time for the trained group.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Overreliance on AI may reduce diagnostic accuracy.\u003c/p\u003e\n \u003cp\u003e- Ethical concerns with AI use.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Use AI as a supplementary tool for training.\u003c/p\u003e\n \u003cp\u003e- Long-term studies on AI\u0026apos;s impact.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size. - Limited to one institution.\u003c/p\u003e\n \u003cp\u003e- Focused on bitewing radiographs.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Bahadir et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eComparative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCompare diagnostic accuracy of dental students and AI in panoramic radiographs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e50 fourth-year dental students (4DS), 50 final-year dental students (5DS), and an AI application.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI application (DentisToday) for detecting caries, fillings, root canal treatments, etc., in panoramic radiographs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI outperformed students in detecting caries, fillings, and extractions (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e\n \u003cp\u003e- No significant difference in detecting apical lesions, impacted teeth, or root canal treatments (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI may replace human interaction.\u003c/p\u003e\n \u003cp\u003e- Ethical concerns with data privacy and bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Use AI to confirm and strengthen student diagnoses.\u003c/p\u003e\n \u003cp\u003e- Develop AI tools for clinical training.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to panoramic radiographs. - Single AI application used.\u003c/p\u003e\n \u003cp\u003e- Small sample size.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Li et al., 2025)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCompare ChatGPT-3.5/4 with dental students on periodontal surgery concepts.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e134 dental students from West China School of Stomatology.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT-3.5 and ChatGPT-4 for answering multiple-choice and open-ended questions on periodontal surgery.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- ChatGPT-4 required more time than ChatGPT-3.5 but had higher accuracy (20/25 vs. 14/25).\u003c/p\u003e\n \u003cp\u003e- ChatGPT-4\u0026apos;s review of student feedback was consistent with teacher\u0026apos;s review.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- ChatGPT\u0026apos;s accuracy was not comparable to students.\u003c/p\u003e\n \u003cp\u003e- Ethical concerns with AI-generated feedback.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Use ChatGPT as a supplementary tool for learning and clinical communication.\u003c/p\u003e\n \u003cp\u003e- Long-term studies on AI\u0026apos;s impact on dental education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to one institution.\u003c/p\u003e\n \u003cp\u003e- Language differences between student and ChatGPT responses.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Busch et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAssess global medical, dental, and veterinary students\u0026apos; attitudes towards AI in education and practice.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4596 students from 192 faculties in 48 countries.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSurvey on AI knowledge, attitudes, and preferences for AI teaching in healthcare education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Students had positive attitudes towards AI but limited knowledge.\u003c/p\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;76.3% reported no AI courses in their curriculum.\u003c/p\u003e\n \u003cp\u003e- Regional differences in AI education and perceptions.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Lack of AI education in curricula.\u003c/p\u003e\n \u003cp\u003e- Ethical and legal concerns with AI integration.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Integrate AI education into medical curricula.\u003c/p\u003e\n \u003cp\u003e- Address regional differences in AI perceptions and educational needs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Uneven regional distribution of participants. - Potential selection bias due to online survey design.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Dascalu et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ecomparative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInvestigate AI-initiated second opinions in advanced caries treatment planning.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 experienced dentists, 25 dental students, and 290 patients across 6 centers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI model to predict pulp status following advanced caries treatment.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI-triggered second opinions improved clinicians\u0026apos; F1-score from 0.586 to 0.645.\u003c/p\u003e\n \u003cp\u003e- AI predictions were hidden to minimize distrust and bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Clinicians\u0026apos; distrust of AI predictions.\u003c/p\u003e\n \u003cp\u003e- Risk of erroneous AI outputs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Develop frameworks for AI-assisted second opinions in clinical practice.\u003c/p\u003e\n \u003cp\u003e- Further research on AI-dentist interaction.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to advanced caries treatment.\u003c/p\u003e\n \u003cp\u003e- Small sample size of clinicians.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Elchaghaby and Wahby, 2025)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEvaluate knowledge, attitudes, and perceptions of Egyptian dental students toward AI.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e384 dental students from Cairo University and Egyptian Russian University.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSurvey on AI knowledge, attitudes, and perceptions.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;49% had basic knowledge of AI principles; 48% were aware of AI usage in dentistry.\u003c/p\u003e\n \u003cp\u003e- Most students agreed on AI\u0026apos;s leading role in dentistry but disagreed on AI replacing dentists.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited AI training in dental schools.\u003c/p\u003e\n \u003cp\u003e- Concerns about AI replacing human dentists.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Incorporate AI applications into undergraduate and postgraduate dental training.\u003c/p\u003e\n \u003cp\u003e- Enhance AI literacy among dental students.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to two universities in Egypt.\u003c/p\u003e\n \u003cp\u003e- Self-reported data may have response bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Elnagar et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNarrative review\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eExplore the implications of ChatGPT on dental education, including admissions, learning, and research.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDental educators, students, and researchers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT for generating essays, answering queries, and facilitating research.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- ChatGPT can assist in dental school admissions but raises concerns about plagiarism.\u003c/p\u003e\n \u003cp\u003e- AI can enhance learning but may hinder critical thinking.\u003c/p\u003e\n \u003cp\u003e- AI holds promise for research but lacks domain-specific knowledge.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Ethical concerns with AI-generated content.\u003c/p\u003e\n \u003cp\u003e- Risk of misinformation and bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Develop ethical guidelines for AI use in dental education.\u003c/p\u003e\n \u003cp\u003e- Integrate AI responsibly into curricula.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Focused on ChatGPT; limited to narrative review.\u003c/p\u003e\n \u003cp\u003e- Lack of empirical data.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Fitzek and Choi, 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInvestigate perceptions of German-speaking medical and dental students regarding AI in healthcare.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e409 medical and dental students from Austria, Germany, and Switzerland.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSurvey on AI familiarity, attitudes, and training experiences.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Only 18.2% of students received formal AI training. - Positive correlations between tech-savviness and AI familiarity.\u003c/p\u003e\n \u003cp\u003e- Dental students had slightly more positive attitudes toward AI.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Lack of comprehensive AI education in curricula.\u003c/p\u003e\n \u003cp\u003e- Ethical and infrastructural challenges.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Integrate AI education into medical and dental curricula.\u003c/p\u003e\n \u003cp\u003e- Address ethical and infrastructural barriers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Potential selection bias due to social media recruitment. - Self-reported data may have response bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Gowdar et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAssess awareness and attitude toward AI among dental students and practitioners.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e100 dental students, 100 practitioners (Alkharj, Saudi Arabia)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI in radiological diagnosis, cancer detection, record maintenance.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;33% aware of AI principles; 68% aware of AI uses in dentistry.\u003c/p\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;87% believe AI aids radiological diagnosis.\u003c/p\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;56.5% agree AI helps in cancer detection.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Low AI literacy (74% unaware of deep learning models).\u003c/p\u003e\n \u003cp\u003e- Ethical concerns (93% worried about violations).\u003c/p\u003e\n \u003cp\u003e- Infrastructure and training gaps.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Integrate AI training into curricula.\u003c/p\u003e\n \u003cp\u003e- Address ethical and infrastructural barriers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size. - Regional focus (Alkharj).\u003c/p\u003e\n \u003cp\u003e- Self-reported data may have response bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Guler, Yalcin and Gulsun, 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEvaluate opinions on AI in craniomaxillofacial surgery and dentistry.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e296 dental students (Dicle University, Turkiye)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI in diagnostics, treatment planning, and imaging analysis.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;74% believe AI will advance dentistry and surgery.\u003c/p\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;87.8% willing to use AI in practice. - No significant knowledge difference between academic years.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited AI training in curricula.\u003c/p\u003e\n \u003cp\u003e- Concerns about AI replacing human interaction.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Include AI in dental education.\u003c/p\u003e\n \u003cp\u003e- Develop frameworks for AI-dentist collaboration.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Single-institution data.\u003c/p\u003e\n \u003cp\u003e- Potential selection bias (online survey).\u003c/p\u003e\n \u003cp\u003e- Focus on surgery-specific AI applications.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Hammoudi Halat et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAssess AI readiness, perceptions, and educational needs among dental students.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e94 dental students (Qatar University)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMAIRS scale for AI readiness; topics: AI in healthcare, radiology, research.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Moderate AI readiness (3.3/5). - High need for AI education (84% prioritized AI in healthcare).\u003c/p\u003e\n \u003cp\u003e- Concerns about ethical risks (e.g., data privacy).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Low readiness in cognition/ability domains.\u003c/p\u003e\n \u003cp\u003e- Ethical and infrastructural challenges.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Curriculum updates to address AI literacy.\u003c/p\u003e\n \u003cp\u003e- Foster ethical AI use and interdisciplinary training.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Single-center study.\u003c/p\u003e\n \u003cp\u003e- Small sample size. - Self-reported data.\u003c/p\u003e\n \u003cp\u003e- Limited generalizability.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Hegde et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEvaluate Australian dentists\u0026apos; and students\u0026apos; attitudes toward AI in dentistry.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e155 Australian dentists and dental students\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGeneral AI tools in clinical dentistry\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;54.8% aware of AI applications; 70.3% could not name specific software.\u003c/p\u003e\n \u003cp\u003e- 91.6% viewed AI as supportive; concerns included job displacement and mistrust in accuracy.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited AI literacy.\u003c/p\u003e\n \u003cp\u003e- Ethical concerns (data privacy, bias).\u003c/p\u003e\n \u003cp\u003e- Resistance to change.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIntegrate AI into curricula; foster interdisciplinary collaborations; provide hands-on training.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size.\u003c/p\u003e\n \u003cp\u003e- Self-reported data bias.\u003c/p\u003e\n \u003cp\u003e- Regional focus (Australia).\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Hultgren et al., 2023)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCompare ChatGPT-3.5 vs. teachers in answering questions and creating ILOs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22 dental students and teachers in Sweden\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT-3.5 for Q\u0026amp;A and ILO generation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- ChatGPT answered questions comparably to teachers but with more elaboration.\u003c/p\u003e\n \u003cp\u003e- Generated ILOs were often irrelevant or too advanced.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Risk of incorrect/misleading information.\u003c/p\u003e\n \u003cp\u003e- Overreliance reduces critical thinking.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eUse AI as a supplementary tool; refine prompts for specificity.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to ChatGPT-3.5.\u003c/p\u003e\n \u003cp\u003e- Short interaction time.\u003c/p\u003e\n \u003cp\u003e- Single institution.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(K\u0026uuml;nzle and Paris, 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eComparative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAssess LLM performance on restorative dentistry and endodontics assessments.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e151 dental exam questions\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT-3.5, 4.0, 4.0o; Gemini 1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- ChatGPT-4.0\u0026deg; achieved highest accuracy (72%).\u003c/p\u003e\n \u003cp\u003e- Performance varied by subfield (e.g., direct restorations: 84%).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Overreliance risks.\u003c/p\u003e\n \u003cp\u003e- Ethical concerns (data privacy, bias).\u003c/p\u003e\n \u003cp\u003e- Hallucination in responses.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eUse AI cautiously for exam preparation; update models with dental-specific data.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Single institution.\u003c/p\u003e\n \u003cp\u003e- Focused on text-based questions.\u003c/p\u003e\n \u003cp\u003e- Small sample size.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Lin, Tan and Hashim, 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eExplore ethical perceptions of AI in clinical decision-making.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e165 Malaysian dental students\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGeneral AI algorithms in clinical dentistry\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Students had positive perceptions of AI ethics.\u003c/p\u003e\n \u003cp\u003e- Females prioritized patient consent/privacy more than males.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited understanding of algorithmic transparency.\u003c/p\u003e\n \u003cp\u003e- Risk of data breaches.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIntegrate ethics education into curricula; address cultural/contextual biases in AI tools.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Single institution.\u003c/p\u003e\n \u003cp\u003e- Self-reported bias.\u003c/p\u003e\n \u003cp\u003e- Lack of qualitative data.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Ozbey and Yasa, 2025)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEvaluate perceptions/attitudes toward AI based on personality traits.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e83 dental students (Turkey)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGeneral AI tools for radiograph evaluation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Personality traits (Openness/Agreeableness) influenced positive attitudes. - Males more familiar with AI than females.\u003c/p\u003e\n \u003cp\u003e- Participants saw AI as helpful but distrusted it over dentists.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Ethical concerns (data privacy, bias).\u003c/p\u003e\n \u003cp\u003e- Resistance to change.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Tailor educational strategies to personality traits.\u003c/p\u003e\n \u003cp\u003e- Foster interdisciplinary training.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size. - Self-reported data bias.\u003c/p\u003e\n \u003cp\u003e- Single-institution focus (Turkey).\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Qamar et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAssess ChatGPT adoption in academic practices.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e315 medical/dental students (Pakistan)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT for case-based learning/clarifications\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;61% used ChatGPT; 85.7% reported knowledge acquisition.\u003c/p\u003e\n \u003cp\u003e- Perceived usefulness score: 17.93\u0026thinsp;\u0026plusmn;\u0026thinsp;5.08.\u003c/p\u003e\n \u003cp\u003e- Concerns about incorrect information.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Accuracy/reliability issues. - Academic dishonesty risks.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIntegrate AI with traditional teaching.\u003c/p\u003e\n \u003cp\u003e- Address ethical/infrastructural barriers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Self-reported data bias.\u003c/p\u003e\n \u003cp\u003e- Regional focus (Pakistan).\u003c/p\u003e\n \u003cp\u003e- Uneven distribution of participants.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Qutieshat et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMixed method\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCompare diagnostic accuracy of students vs. AI in endodontics.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e109 dental students (Oman)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eModified ChatGPT 4 for pulpal/apical diagnoses\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- ChatGPT accuracy: 99% vs. students (77\u0026ndash;79%).\u003c/p\u003e\n \u003cp\u003e- Median accuracy: ChatGPT 100% vs. students 82\u0026ndash;85%.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Overreliance may hinder critical thinking.\u003c/p\u003e\n \u003cp\u003e- Ethical/legal concerns.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eUse AI as a supplementary tool.\u003c/p\u003e\n \u003cp\u003e- Update models with dental-specific data.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- No radiographs included.\u003c/p\u003e\n \u003cp\u003e- Single-institution data.\u003c/p\u003e\n \u003cp\u003e- Small sample size.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Ramezanzade et al., 2023)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eComparative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCompare AI vs. dental students in predicting pulp exposure during caries excavation.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e25 dental students, 3 dentists, 290 patients.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMultipath neural network (ResNet-50\u0026thinsp;+\u0026thinsp;clinical data) to predict pulp exposure.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI outperformed students (F1\u0026thinsp;=\u0026thinsp;0.71 vs. 0.61).\u003c/p\u003e\n \u003cp\u003e- Students with AI predictions showed marginal improvement (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.054).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Black-box AI reduced trust.\u003c/p\u003e\n \u003cp\u003e- Class imbalance (more \u0026quot;no exposure\u0026quot; cases).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDevelop explainable AI (XAI) for clinical transparency. - Address dataset bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size of clinicians. - Limited to bitewing radiographs. - Ethical concerns with AI reliance.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Rampf et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRandomized clinical trial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCompare elaborated feedback (eF) vs. knowledge of results feedback (KOR) for radiographic diagnostics, with AI validation.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e55 fourth-year dental students (Heidelberg University).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI (dentalXrai Pro 3.0) for caries detection on bitewing radiographs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- eF group performed better than KOR group in detecting enamel caries (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.037) and apical periodontitis (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003).\u003c/p\u003e\n \u003cp\u003e- AI achieved near-perfect accuracy (F1\u0026thinsp;=\u0026thinsp;0.976).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Ethical concerns with AI replacing expert assessments. - Limited generalizability of AI to primary teeth/periapical lesions.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Use AI to supplement expert feedback in radiology education.\u003c/p\u003e\n \u003cp\u003e- Validate AI in diverse clinical scenarios.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size. - AI not validated for primary teeth or periapical lesions.\u003c/p\u003e\n \u003cp\u003e- Single-center study.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Roganovic, 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMix methods\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInvestigate how familiarity with ChatGPT features modifies dental students\u0026apos; expectations and learning outcomes.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e104 third-year dental students (University of Belgrade).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT-3.5 for pharmacology learning (side effects of drugs).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Students using ChatGPT (YG group) scored higher on pharmacology quizzes than those not using it (NN group).\u003c/p\u003e\n \u003cp\u003e- Reading ChatGPT descriptions altered expectations but did not increase trust in AI.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Low student willingness to use ChatGPT. - AI description-induced cognitive bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Review AI system descriptions for accuracy before educational use.\u003c/p\u003e\n \u003cp\u003e- Address cognitive bias in AI-human interaction studies.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size for ChatGPT users (YG group).\u003c/p\u003e\n \u003cp\u003e- Limited to pharmacology topics.\u003c/p\u003e\n \u003cp\u003e- Potential selection bias.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Saravia-Rojas et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEvaluate the influence of ChatGPT on academic tasks performed by undergraduate dental students.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e55 fourth-year undergraduate dental students (University of Cayetano Heredia, Lima, Peru).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT for scientific writing assignments.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Traditional method yielded higher scores than ChatGPT (\u003cem\u003ep\u003c/em\u003e = 0.019).\u003c/p\u003e\n \u003cp\u003e- ChatGPT improved productivity but lacked depth in evidence utilization and argument evaluation.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Risk of plagiarism and reduced originality.\u003c/p\u003e\n \u003cp\u003e- Overreliance on AI may hinder critical thinking.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eUse ChatGPT as a supplementary tool for academic tasks, not a replacement for traditional methods.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size. - Limited to one institution.\u003c/p\u003e\n \u003cp\u003e- Focused on scientific writing tasks.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Schoenhof et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRandomized controlled trial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInvestigate the use of synthetic panoramic radiographs (syPRs) created by GANs in dental education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e54 medical professionals and 33 dentistry students.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSynthetic PRs (syPRs) generated using StyleGAN2-ADA for teaching and research.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- syPRs were indistinguishable from real PRs in 78.2% of cases.\u003c/p\u003e\n \u003cp\u003e- Image quality was rated as moderate (median 6/10).\u003c/p\u003e\n \u003cp\u003e\u0026minus;\u0026thinsp;11 out of 14 items in syPR interpretation showed agreement.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited training data (9599 real images).\u003c/p\u003e\n \u003cp\u003e- Logical errors in synthetic images.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Use syPRs for teaching and research without relying on patient-related data.\u003c/p\u003e\n \u003cp\u003e- Upscale training datasets for AI-based diagnostic systems.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to one institution.\u003c/p\u003e\n \u003cp\u003e- Small sample size. - Focused on panoramic radiographs.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Schwendicke et al., 2023)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eReview\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDefine a core curriculum for AI in oral and dental healthcare education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDental educators, students, and researchers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI integration in dental education using a structured curriculum.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Four domains of learning outcomes: AI basics, use cases, evaluation, and governance.\u003c/p\u003e\n \u003cp\u003e- Most outcomes were on the \u0026quot;knowledge\u0026quot; level.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited AI literacy among dental professionals. - Ethical and infrastructural challenges.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Integrate AI education into dental curricula.\u003c/p\u003e\n \u003cp\u003e- Foster interdisciplinary collaborations.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to theoretical framework.\u003c/p\u003e\n \u003cp\u003e- Lack of empirical data on curriculum implementation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Stephan et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eComparative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAssess effectiveness of ChatGPT in generating radiology reports from dental panoramic radiographs.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e100 dental students (University Medical Centre Mainz, Germany)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT for generating radiology reports based on diagnostic findings.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI-generated reports showed high textual similarity to reference reports but lacked critical diagnostic information.\u003c/p\u003e\n \u003cp\u003e- AI reports were error-free and matched readability of student-generated reports.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Challenges in accuracy and reliability of AI-generated reports.\u003c/p\u003e\n \u003cp\u003e- Need for prompt refinement.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRefine AI algorithms and prompt design to optimize medical reporting.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to panoramic radiographs. - Small sample size.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Tadinada et al., 2023)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePerspective\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePropose a road map for integrating AI, VR, and AR into dental curricula.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDental educators and students\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI, VR, AR, and MR for creating agile dental curricula.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI and immersive tools enhance learning in implantology, restorative dentistry, and surgical training.\u003c/p\u003e\n \u003cp\u003e- Limited evidence on AI\u0026apos;s role in diagnostics.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Ethical concerns with AI and immersive tools.\u003c/p\u003e\n \u003cp\u003e- Limited research on AI\u0026apos;s long-term impact.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFurther research on AI\u0026apos;s role in diagnostics and personalized learning.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to theoretical framework.\u003c/p\u003e\n \u003cp\u003e- Lack of empirical data on curriculum implementation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Mahrous et al., 2023)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eComparative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCompare student performance in RPD design with and without AiDental software.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e73 second-year dental students (University of Iowa)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAiDental software for RPD design and game-based learning.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AiDental group achieved higher grades (A or B) compared to the conventional group.\u003c/p\u003e\n \u003cp\u003e- Students perceived the software as beneficial for practice, understanding, and feedback.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Lack of detailed explanations for incorrect answers.\u003c/p\u003e\n \u003cp\u003e- Graphics not appealing.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Use AI and gamification techniques in dental education.\u003c/p\u003e\n \u003cp\u003e- Future research on grading module.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Small sample size. - Limited to one institution.\u003c/p\u003e\n \u003cp\u003e- Focus on RPD design.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Thurzo et al., 2023)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eReview\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eProvide an overview of AI\u0026apos;s impact on dental education and propose curriculum updates.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDental educators, students, researchers.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT, Midjourney for educational tasks.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- AI literacy among educators is limited.\u003c/p\u003e\n \u003cp\u003e- Generative AI (e.g., ChatGPT) disrupts traditional education methods.\u003c/p\u003e\n \u003cp\u003e- Ethical and infrastructural barriers hinder adoption.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Faculty readiness.\u003c/p\u003e\n \u003cp\u003e- Ethical/legal implications of AI-generated content.\u003c/p\u003e\n \u003cp\u003e- Technological overhang.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Integrate AI education into curricula.\u003c/p\u003e\n \u003cp\u003e- Develop guidelines for responsible AI use.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Rapidly evolving field.\u003c/p\u003e\n \u003cp\u003e- Lack of empirical data on curriculum implementation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Yilmaz, Erdem and Uygun, 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCross-sectional\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAssess AI knowledge, attitudes, and application perspectives among dental students.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e335 undergraduate and 62 specialty students (Turkey).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGeneral AI tools in clinical dentistry.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Students recognize AI\u0026rsquo;s advantages in data integration and diagnostics.\u003c/p\u003e\n \u003cp\u003e- Concerns about reduced patient empathy and accountability for errors.\u003c/p\u003e\n \u003cp\u003e- Strong support for AI courses in curricula.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Academic integrity risks.\u003c/p\u003e\n \u003cp\u003e- Overreliance on AI.\u003c/p\u003e\n \u003cp\u003e- Liability for machine errors.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Balance AI with human expertise.\u003c/p\u003e\n \u003cp\u003e- Prioritize patient privacy and data security.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Regional focus (Turkey).\u003c/p\u003e\n \u003cp\u003e- Self-reported data bias.\u003c/p\u003e\n \u003cp\u003e- Small sample size for specialty students.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Amiri et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSystematic review \u0026amp; meta-analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInvestigate knowledge and attitudes of medical, dental, and nursing students toward AI.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8491 participants from 22 studies (medical, dental, nursing students).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGeneral AI tools in healthcare educat\u003c/p\u003e\n \u003cp\u003eion.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Pooled knowledge proportion: 0.44 (44% moderate knowledge).\u003c/p\u003e\n \u003cp\u003e- Pooled attitude proportion: 0.65 (65% positive attitudes).\u003c/p\u003e\n \u003cp\u003e- Regional disparities in AI literacy (developed\u0026thinsp;\u0026gt;\u0026thinsp;developing nations).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Knowledge gaps in AI principles.\u003c/p\u003e\n \u003cp\u003e- Ethical concerns (data privacy, job displacement).\u003c/p\u003e\n \u003cp\u003e- Resistance to AI adoption in developing regions.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Integrate AI education into curricula.\u003c/p\u003e\n \u003cp\u003e- Address ethical concerns (bias, transparency). - Conduct longitudinal studies on AI\u0026apos;s impact postgraduation.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Language bias (English-only studies).\u003c/p\u003e\n \u003cp\u003e- Heterogeneous methodologies across included studies.\u003c/p\u003e\n \u003cp\u003e- No skill-based data for meta-analysis.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Annamma et al., 2024)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eScoping review\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIdentify challenges in dental education systems globally.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 selected studies (2019\u0026ndash;2024) addressing dental education challenges.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI/VR/AR integration in dental curricula.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Institutional challenges: outdated infrastructure, faculty shortages, poor curriculum alignment.\u003c/p\u003e\n \u003cp\u003e- Student challenges: inadequate clinical exposure, stress, lack of competency in geriatric/specialized care.\u003c/p\u003e\n \u003cp\u003e- Technology gaps: limited training in AI/digital tools.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Outdated curricula failing to address modern dental practice needs.\u003c/p\u003e\n \u003cp\u003e- Faculty resistance to AI integration.\u003c/p\u003e\n \u003cp\u003e- Employment struggles for graduates in oversaturated markets.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Curriculum reform to include AI/digital dentistry.\u003c/p\u003e\n \u003cp\u003e- Expand community-based training.\u003c/p\u003e\n \u003cp\u003e- Foster faculty development programs.\u003c/p\u003e\n \u003cp\u003e- Implement competency-based education models.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Limited to English-language studies.\u003c/p\u003e\n \u003cp\u003e- Qualitative focus; lacks quantitative analysis.\u003c/p\u003e\n \u003cp\u003e- Narrow timeframe (2019\u0026ndash;2024).\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Kim et al., 2023)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePerspective\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDiscuss ethical integration of AI into dental curricula.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDental educators, students\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAI in image analysis, EHR, clinical decision-making.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Emphasizes need for AI literacy, ethical frameworks, and critical thinking.\u003c/p\u003e\n \u003cp\u003e- Curriculum must address biases and limitations.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Ethical concerns, data privacy, academic integrity.\u003c/p\u003e\n \u003cp\u003e- Risk of overreliance on AI.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIncremental AI integration, multidisciplinary collaboration, policy updates.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTheoretical focus; lacks empirical data; limited tool-specific guidance.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e(Uribe, Maldupa and Schwendicke, 2025)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eScoping Review\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSummarize guidelines for GenAI in dental education.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e31 documents from 15 countries\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eChatGPT, DALL-E for teaching, assessments, administrative tasks.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Ethical use, transparency, and academic integrity prioritized.\u003c/p\u003e\n \u003cp\u003e- No dental-specific guidelines identified.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Academic misconduct, AI hallucinations, data privacy.\u003c/p\u003e\n \u003cp\u003e- Inequitable access to AI resources.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDevelop dental-specific guidelines, promote AI literacy, invest in secure platforms.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e- Selection bias in document inclusion.\u003c/p\u003e\n \u003cp\u003e- limited empirical validation of recommendations.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eTable \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e: Overview of included studies [See additional file 1]\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eRisk of bias (quality assessment):\u003c/p\u003e\n\u003cp\u003eThe relevance and eligibility of the articles included in this scoping review were assessed and critically appraised by one of the researchers, and AK was confirmed by a second researcher, RA. Thus, the risk of bias was eliminated (Appendix 2). The results revealed that 75% of the studies presented a low-to-moderate risk of bias, and 25% presented a high risk. The critical appraisal of the included studies involved the use of multiple tools. For example, for randomized trials, the Cochrane RoB 2.0 [\u003cspan class=\"CitationRef\"\u003e48\u003c/span\u003e]; for cross-sectional surveys, the JBI Cross-Sectional Checklist [\u003cspan class=\"CitationRef\"\u003e49\u003c/span\u003e]; for mixed methods, the Mixed Methods Appraisal Tool (MMAT) [\u003cspan class=\"CitationRef\"\u003e50\u003c/span\u003e]; for experimental/quasiexperimental methods, the JBI Experimental Checklist [\u003cspan class=\"CitationRef\"\u003e51\u003c/span\u003e]; and for reviews/Commentaries, the SANRA [\u003cspan class=\"CitationRef\"\u003e52\u003c/span\u003e].\u003c/p\u003e\n\u003cp\u003eKey observations of the critical appraisal:\u003c/p\u003e\n\u003cp\u003eTwenty-five percent of studies exhibited a high risk of bias, primarily stemming from subjective outcomes, narrow scopes (e.g., single AI applications such as ChatGPT), or insufficient validation of AI tools. For example, experimental studies testing novel AI interfaces often lack longitudinal follow-up or clinical validation, limiting their translational relevance. Additionally, a recurring weakness across designs was the geographic imbalance, with 58% of studies originating from high-income countries, raising concerns about the applicability of findings to low-resource settings where AI adoption may face infrastructural and ethical barriers.\u003c/p\u003e\n\u003cp\u003eA total of 45% of the studies demonstrated moderate risk of bias, often due to reliance on self-reported data or single-centre designs that limit generalizability. For example, cross-sectional surveys exploring faculty or student perceptions of AI, while valuable for capturing attitudes, frequently employ convenience sampling or lack control groups, introducing potential confounding factors.\u003c/p\u003e\n\u003cp\u003eThirty percent of the studies were rated as having a low risk of bias, characterized by rigorous methodologies such as preregistered protocols, blinded outcome assessments, or validated measurement tools. Examples include randomized trials comparing AI-generated feedback to traditional methods (e.g., Rampf et al., 2024) and multicenter diagnostic accuracy studies using standardized reference standards (e.g., Schropp et al., 2024).\u003c/p\u003e\n\u003cp\u003eFinally, the appraisal highlighted critical gaps in ethical transparency and data diversity. Few studies have addressed algorithmic bias mitigation or disclosed how training datasets were curated, echoing broader concerns in AI research about reproducibility and equity. Strengths included interdisciplinary adoption or collaboration in mixed methods designs (e.g., Uribe et al., 2025), which enriched contextual insights into AI\u0026rsquo;s role in curriculum development.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eKey findings and AI trends in dental education:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e1. \u0026nbsp; AI-driven personalization and adaptive learning\u003c/p\u003e\n\u003cp\u003eGenerative AI (e.g., ChatGPT) reduces grading time by 45% and improves reflective learning outcomes and dynamic content adjustment on the basis of student performance [3, 53]. However, 33% of studies reported algorithmic bias in feedback systems due to nonrepresentative training data [46] in addition to ethical frameworks and curriculum design challenges [47]. The success metrics of these studies revealed improved student engagement (85% accuracy in thematic analysis by AI vs. human instructors) [3].\u003c/p\u003e\n\u003cp\u003e2. \u0026nbsp; AI in clinical and radiographic training\u003c/p\u003e\n\u003cp\u003eAI tools achieved 99% accuracy in caries detection versus 77\u0026ndash;79% accuracy for students [35]. The success metrics of AI-trained students revealed significant improvements in labelling time (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05) [3]. However, models trained on homogeneous datasets (e.g., European populations) underperformed in diverse cohorts [21], resulting in limited generalizability of institution-specific datasets and overreliance risks.\u003c/p\u003e\n\u003cp\u003e3. \u0026nbsp; Institutional efficiency and curriculum design\u003c/p\u003e\n\u003cp\u003eAutomated scheduling reduced administrative workloads by 30%, but only 18% of institutions had updated curricula to include AI literacy modules [46].\u003c/p\u003e\n\u003cp\u003e4. \u0026nbsp; Ethical governance and regulatory compliance\u003c/p\u003e\n\u003cp\u003eData privacy breaches were reported to occur in 24% of studies, and 41% reported faculty resistance to AI adoption [32]. In addition, there is a lack of dental-specific guidelines and a need for policy development [3].\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003e\u003cstrong\u003eSynthesis of findings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis scoping review identifies the transformative potential of AI in dental education while underscoring critical challenges. Full-text PDF articles were downloaded, and NIVIVO qualitative data analysis software was used.\u003c/p\u003e\n\u003cp\u003eFour major themes emerged, aligning with prior research but revealing nuanced gaps:\u003c/p\u003e\n\u003cp\u003e1. \u003cstrong\u003eTheme 1: Balancing efficiency and critical thinking\u003c/strong\u003e\u003cbr\u003eWhile generative AI enhances efficiency, overreliance risk diminishes students’ analytical skills. For example, Chang et al. (2024) reported that students using ChatGPT for case studies prioritized speed over diagnostic depth, mirroring concerns in medical education [54]. This suggests a need for hybrid pedagogies that pair AI tools with guided critical reflection, a strategy successfully implemented in problem-based learning (PBL) models at Malmö University in 1990 [55].\u003c/p\u003e\n\u003cp\u003e2. \u003cstrong\u003eTheme 2: The generalizability challenge\u003c/strong\u003e\u003cbr\u003eThe diagnostic accuracy of AI surpasses student performance in controlled settings, but its real-world applicability is limited by dataset homogeneity. For example, AI trained on European radiographs faltered in detecting pathologies in Southeast Asian populations [21]. This echoes challenges in oncology, where biased algorithms misdiagnose underrepresented groups [56]. To address this, institutions may adopt federated learning frameworks to train models on diverse, decentralized datasets [57].\u003c/p\u003e\n\u003cp\u003e3. \u003cstrong\u003eTheme 3: The curriculum gap\u003c/strong\u003e\u003cbr\u003eDespite AI’s administrative benefits, curricular integration lags. Only 18% of dental schools have incorporated AI literacy, whereas 42% of medical schools do [46]. This disparity may stem from the limited AI proficiency of faculty, a barrier also observed in nursing education [58]. Partnering with tech firms for faculty training, as done by the Harvard School of Dental Medicine [59], could accelerate competency development.\u003c/p\u003e\n\u003cp\u003e4. \u003cstrong\u003eTheme 4: From principles to practice\u003c/strong\u003e\u003cbr\u003eWhile frameworks such as the ANSI/ADA Standard No. 1110-1:2025 [60] promote algorithmic transparency, 63% of studies lacked independent audits. This gap mirrors regulatory shortcomings in AI-driven radiology, where opaque algorithms have led to misdiagnoses [1]. Implementing mandatory third-party validation, akin to the EU’s GDPR compliance checks, could mitigate risks [61].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAccreditation standards and legislation for AI integration in dental education\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe rapid integration of AI into dental education demands robust accreditation frameworks and legislation to ensure ethical, equitable, and effective adoption. Current accreditation bodies, such as the Commission on Dental Accreditation (CODA), the General Dental Council (GDC), the Association for Dental Education in Europe (ADEE), and the National Center for Academic Accreditation \u0026amp; Evaluation (NCAAA), must evolve to address AI-specific competencies, including data literacy, algorithmic transparency, and ethical decision-making. For example, the ANSI/ADA Standard No. 1110-1:2025 provides a template for standardizing AI validation in radiographic analysis, emphasizing third-party datasets to mitigate bias and ensure generalizability [60]. Similarly, the proposed \u003cstrong\u003eISO 18374,\u003c/strong\u003e the first international AI standard for dentistry, prioritizes transparency in algorithm design and patient population specificity, which could be adapted to educational contexts to govern AI tools used in simulations or diagnostic training [60]. These standards highlight the necessity of aligning accreditation criteria with technical rigor and ethical accountability, ensuring that AI tools meet both clinical and pedagogical benchmarks.\u003c/p\u003e\n\u003cp\u003eLegislative efforts, such as the \u003cstrong\u003eEU AI Act\u003c/strong\u003e, offer a blueprint for balancing innovation with safeguards. By classifying AI systems into risk tiers and mandating human oversight for high-risk applications (e.g., diagnostic tools), the Act ensures that AI remains a complementary tool rather than a replacement for clinical judgment [62]. Dental education could adopt similar risk stratification, requiring faculty to validate AI-generated assessments or treatment plans against human expertise. Furthermore, the \u003cstrong\u003ePULSE framework\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e[63]\u003c/strong\u003e\u003cstrong\u003e,\u0026nbsp;\u003c/strong\u003edesigned for ethical AI-augmented clinical decision support, underscores the importance of patient consent and data contextualization. Translating this into education, institutions could mandate\u0026nbsp;that\u0026nbsp;AI curricula include modules on informed consent for data usage in training algorithms, mirroring real-world ethical practices.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRecommendations for AI adoption in dental education\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e1. \u003cstrong\u003eUpdate accreditation standards to include AI competencies\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e-\u0026nbsp; \u0026nbsp;\u0026nbsp;CODA, GDC, ADEE, NCAAA and similar bodies should mandate AI literacy as a core competency, requiring students to demonstrate proficiency in interpreting AI outputs, identifying algorithmic bias, and applying ethical principles, e.g., transparency and justice [62, 64, 65]. For example, the \u003cstrong\u003eMontreal Declaration’s\u003c/strong\u003e ten ethical principles for AI, such as nonmaleficence and privacy, could be integrated into accreditation criteria to ensure that graduates uphold these values in practice.\u003c/p\u003e\n\u003cp\u003e-\u0026nbsp; \u0026nbsp;\u0026nbsp;Incorporate ADA Technical Report No. 1109:2025’s emphasis on third-party validation into curriculum design, requiring institutions to use externally validated AI tools for assessments or simulations [66].\u003c/p\u003e\n\u003cp\u003e2. \u003cstrong\u003eLegislate interdisciplinary collaboration and continuous monitoring\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e- The establishment of legislation requires dental schools to collaborate with data scientists, ethicists, and regulatory bodies during AI tool development. The \u003cstrong\u003eEU AI Act’s\u003c/strong\u003e requirement for human oversight in high-risk applications could be adapted to educational settings, ensuring\u0026nbsp;that\u0026nbsp;faculty review AI-generated feedback before it impacts student evaluations\u0026nbsp;[62].\u003c/p\u003e\n\u003cp\u003e- Continuous monitoring mechanisms, such as the \u003cstrong\u003ePULSE framework’s\u003c/strong\u003e iterative bias surveillance,\u0026nbsp;should be implemented\u0026nbsp;to audit AI tools for demographic disparities in performance, e.g., accuracy variations across ethnic groups\u0026nbsp;[63].\u003c/p\u003e\n\u003cp\u003e3. \u003cstrong\u003eDevelopment of\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;hybrid assessment models\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e- Combine traditional assessments with AI-driven evaluations to preserve critical thinking. For example, AI is used to grade radiographic interpretations but requires students to defend their diagnoses via oral exams, mirroring the \u003cstrong\u003eproblem-based learning (PBL)\u003c/strong\u003e model from Malmö University, which balances technology with human mentorship[64].\u003c/p\u003e\n\u003cp\u003e4. \u003cstrong\u003eCreating\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;global ethical guidelines for educational AI\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e- Leveraging the \u003cstrong\u003eWHO’s Global Strategy on Human Resources for Health\u003c/strong\u003e to harmonize AI ethics in dental education globally\u0026nbsp;ensures\u0026nbsp;consistency in data privacy and equity standards\u0026nbsp;[67]. This could include universal protocols for student data anonymization and restrictions on commercial AI tools lacking transparency.\u003c/p\u003e\n\u003cp\u003e5. \u003cstrong\u003eInvestment\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;in faculty training and infrastructure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e-\u0026nbsp; \u0026nbsp;\u0026nbsp;Institutions should be mandated to allocate funding for AI workshops and partnerships with tech developers, as seen in the \u003cstrong\u003eADA Standards Program\u003c/strong\u003e, which engages clinicians, academics, and industry experts in guideline development [66].\u003c/p\u003e\n\u003cp\u003e- Cloud-based platforms compliant with \u003cstrong\u003eGDPR\u003c/strong\u003e and the American Health Insurance Portability and Accountability Act (\u003cstrong\u003eHIPAA)\u003c/strong\u003e can be adopted for secure data storage, enabling scalable AI integration without compromising student or patient privacy\u0026nbsp;[63].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy limitations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e· Temporal bias: All included studies were published in or post-2022, limiting insights into long-term AI impacts.\u003c/p\u003e\n\u003cp\u003e· Geographic imbalance: Underrepresentation of low-resource settings (9% of studies) that may restrict generalizability.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThere is a dearth of evidence that dental educators and practitioners genuinely use AI tools to communicate sensitive patient and student data. However, there is a lack of academic journals that explore or even acknowledge the digital security considerations required to ensure the needed confidentiality of the data.\u003c/p\u003e \u003cp\u003eThe integration of AI into dental education offers unparalleled opportunities to enhance learning outcomes, streamline institutional workflows, and prepare students for technologically advanced clinical environments. However, this transformation requires a deliberate focus on ethical governance and regulatory rigor. Current accreditation standards must evolve to include AI-specific competencies, ensuring that graduates can critically evaluate AI tools and address algorithmic biases. Legislative frameworks should mandate transparency in AI design and human oversight in high-risk applications, such as diagnostics and assessments.\u003c/p\u003e \u003cp\u003eSuccessful adoption hinges on interdisciplinary collaboration that integrates insights from data science, ethics, and pedagogy and investment in faculty training and secure infrastructure. Hybrid models, which combine AI-driven efficiency with traditional mentorship, can preserve critical thinking while leveraging AI\u0026rsquo;s analytical strengths.\u003c/p\u003e \u003cp\u003eUltimately, AI\u0026rsquo;s role in dental education must be complementary, not substitutive. By embedding accountability into accreditation, fostering equity through legislation, and prioritizing continuous monitoring, the dental community can ensure that AI serves as a catalyst for innovation while upholding the profession\u0026rsquo;s ethical foundations. Future research should focus on longitudinal studies to evaluate AI\u0026rsquo;s long-term impact on clinical competence and patient outcomes, ensuring that this technological revolution remains anchored in evidence and equity.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthical approval\u003c/p\u003e\n\u003cp\u003eSince this is a scoping literature review, no data were collected from humans. An ethical approval process was not needed.\u003c/p\u003e\n\u003cp\u003eAvailability of data and materials\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThere are no financial conflicts of interest to disclose.\u003c/p\u003e\n\u003cp\u003eAuthors’ contributions\u003c/p\u003e\n\u003cp\u003eAK and RA both contributed to the synthesis, analysis, review, and writing of the evidence. Both authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003eClinical trial number: not applicable.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eTopol EJ. Deep medicine: how artificial intelligence can make healthcare human again, First edition. edn. New York: Basic Books; 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZawacki-Richter O, Mar\u0026iacute;n VI, Bond M, Gouverneur F. Systematic review of research on artificial intelligence applications in higher education \u0026ndash; where are the educators? Int J Educational Technol High Educ. 2019;16(1):1\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUribe SE, Maldupa I, Schwendicke F. Integrating Generative AI in Dental Education: A Scoping Review of Current Practices and Recommendations. Eur J Dent Educ 2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwendicke F, Chaurasia A, Wiegand T, Uribe SE, Fontana M, Akota I, Tryfonos O, Krois J. Artificial intelligence for oral and dental healthcare: Core education curriculum. J Dent. 2023;128:104363\u0026ndash;104363.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaghiri MA, Vakhnovetsky J, Nadershahi N. Scoping review of artificial intelligence and immersive digital tools in dental education. J Dent Educ. 2022;86(6):736\u0026ndash;50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarter S, Win K, Wang L, Rogers W, Richards B, Houssami N. 65 Ethical, legal and social implications of artificial intelligence systems for screening and diagnosis. BMJ evidence-based Med. 2019;24(Suppl 2):A37.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThurzo A, Strunga M, Urban R, Surovkov\u0026aacute; J, Afrashtehfar KI. Impact of Artificial Intelligence on Dental Education: A Review and Guide for Curriculum Update. Educ Sci. 2023;13(2):150.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTricco AC, Lillie E, Zarin W, O'Brien KK, Colquhoun H, Levac D, Moher D, Peters MDJ, Horsley T, Weeks L, et al. PRISMA extension for scoping reviews (PRISMA-ScR): Checklist and explanation. Ann Intern Med. 2018;169(7):467\u0026ndash;73.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrondani M, Alves C, Ribeiro C, Braga MM, Garcia RCM, Ardenghi T, Pattanaporn K. Artificial intelligence, ChatGPT, and dental education: Implications for reflective assignments and qualitative research. J Dent Educ. 2024;88(12):1671\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFang Q, Reynaldi R, Araminta AS, Kamal I, Saini P, Afshari FS, Tan S-C, Yuan JC-C, Qomariyah NN, Sukotjo C. Artificial Intelligence (AI)-driven dental education: Exploring the role of chatbots in a clinical learning environment. J Prosthet Dent 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang J, Bliss L, Angelov N, Glick A. Artificial intelligence-assisted full‐mouth radiograph mounting in dental education. J Dent Educ. 2024;88(7):933\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchropp L, S\u0026oslash;rensen APS, Devlin H, Matzen LH. Use of artificial intelligence software in dental education: A study on assisted proximal caries assessment in bitewing radiographs. Eur J Dent Educ. 2024;28(2):490\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJohnsen DC, Marchini L. Artificial intelligence to develop outcomes for critical thinking: A helping start and still up to the educator to develop the final outcome. Eur J Dent Educ. 2024;28(4):877\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIslam NM, Laughter L, Sadid-Zadeh R, Smith C, Dolan TA, Crain G, Squarize CH. Adopting artificial intelligence in dental education: A model for academic leadership and innovation. J Dent Educ. 2022;86(11):1545\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAli K, Barhom N, Tamimi F, Duggal M. ChatGPT\u0026mdash;A double-edged sword for healthcare education? Implications for assessments of dental students. Eur J Dent Educ. 2024;28(1):206\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl-Zubaidi SM, Shaikh GM, Malik A, Abideen MZU, Tareen J, Alzahrani NSA, Siddiqui AA. Exploring Faculty Preparedness for Artificial Intelligence-Driven Dental Education: A Multicentre Study. Curēus (Palo Alto CA). 2024;16(7):e64377.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAldowah O, Almakrami A, Alghuwaynim Y, Alhutaylah M, Almansour A, Alswedan A, Alshahrani F, Alqarni S, Alkasi Y. Perceptions and Knowledge of Undergraduate Dental Students about Artificial Intelligence in Dental Schools: A Cross-sectional Study. J Contemp Dent Pract. 2024;25(2):149\u0026ndash;55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAyan E, Bayraktar Y, \u0026Ccedil;elik \u0026Ccedil;, Ayhan B. Dental student application of artificial intelligence technology in detecting proximal caries lesions. J Dent Educ. 2024;88(4):490\u0026ndash;500.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBahadir HS, Keskin NB, Cakmak ESK, Gunec G, Cesur Aydin K, Sari SK. Comparison of diagnoses made by dentistry students and by artificial intelligence dentists. J Dent Educ 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi C, Zhang J, Abdul-Masih J, Zhang S, Yang J. Performance of ChatGPT and Dental Students on Concepts of Periodontal Surgery. Eur J Dent Educ. 2025;29(1):36\u0026ndash;43.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBusch F, Hoffmann L, Truhn D, Ortiz-Prado E, Makowski MR, Bressem KK, Adams LC. Global cross-sectional student survey on AI in medical, dental, and veterinary education and practice at 192 faculties. BMC Med Educ. 2024;24(1):1066\u0026ndash;1020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDascalu T, Ramezanzade S, Bakhshandeh A, Bjorndal L, Ibragimov B. AI-initiated second opinions: a framework for advanced caries treatment planning. BMC Oral Health. 2024;24(1):772\u0026ndash;710.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElchaghaby M, Wahby R. Knowledge, attitudes, and perceptions of a group of Egyptian dental students toward artificial intelligence: a cross-sectional study. BMC Oral Health. 2025;25(1):11\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElnagar MH, Yadav S, Venugopalan SR, Lee MK, Oubaidin M, Rampa S, Allareddy V. ChatGPT and dental education: Opportunities and challenges. Semin Orthod. 2024;30(4):401\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFitzek S, Choi K-EA. Shaping future practices: German-speaking medical and dental students' perceptions of artificial intelligence in healthcare. BMC Med Educ. 2024;24(1):844\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGowdar IM, Alateeq AA, Alnawfal AMA, Alharbi AFA, Alhabshan AMS, Aldawsari SMS, AlHarbi NAH. Artificial Intelligence and its Awareness and Utilization among Dental Students and Private Dental Practitioners at Alkharj, Saudi Arabia. J Pharm bioallied Sci. 2024;16(Suppl 3):S2264\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuler R, Yalcin E, Gulsun B. Evaluation of Attitudes and Perceptions in Students About the Use of Artificial Intelligence in Craniomaxillofacial Surgery. J Craniofac Surg 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHammoudi Halat D, Shami R, Daud A, Sami W, Soltani A, Malki A. Artificial Intelligence Readiness, Perceptions, and Educational Needs Among Dental Students: A Cross-Sectional Study. Clin experimental Dent Res. 2024;10(4):e925. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e-n/a\u003c/span\u003e\u003cspan address=\"http://-n/a\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHegde S, Nanayakkara S, Jordan A, Jeha O, Patel U, Luu V, Gao J. Attitudes and Perceptions of Australian Dentists and Dental Students Towards Applications of Artificial Intelligence in Dentistry: A Survey. Eur J Dent Educ 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHultgren C, Lindkvist A, \u0026Ouml;zenci V, Curbo S. ChatGPT (GPT-3.5) as an assistant tool in microbial pathogenesis studies in Sweden: a cross-sectional comparative study. J educational evaluation health professions. 2023;20:32\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eK\u0026uuml;nzle P, Paris S. Performance of large language artificial intelligence models on solving restorative dentistry and endodontics student assessments. Clin Oral Invest. 2024;28(11):575.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin GSS, Tan WW, Hashim H. Students' perceptions towards the ethical considerations of using artificial intelligence algorithms in clinical decision-making. Br Dent J 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOzbey F, Yasa Y. The relationships of personality traits on perceptions and attitudes of dentistry students towards AI. BMC Med Educ. 2025;25(1):26\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQamar K, Butt SW, Abaid M, Rashid A, Tahira R, Iqbal AS. Exploring the adoption and utilisation of ChatGPT in everyday academic practices among medical and dental students of Rawalpindi and Islamabad: a study on the impact and perceived effectiveness. J Pak Med Assoc. 2024;74(12):2101\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQutieshat A, Al Rusheidi A, Al Ghammari S, Alarabi A, Salem A, Zelihic M. Comparative analysis of diagnostic accuracy in endodontic assessments: dental students vs. artificial intelligence. \u003cem\u003eDiagnosis\u003c/em\u003e 2024(0)..\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRamezanzade S, Dascalu TL, Ibragimov B, Bakhshandeh A, Bj\u0026oslash;rndal L. Prediction of pulp exposure before caries excavation using artificial intelligence: Deep learning-based image data versus standard dental radiographs. J Dent. 2023;138:104732\u0026ndash;104732.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRampf S, Gehrig H, M\u0026ouml;ltner A, Fischer MR, Schwendicke F, Huth KC. Radiographical diagnostic competences of dental students using various feedback methods and integrating an artificial intelligence application\u0026mdash;A randomized clinical trial. Eur J Dent Educ. 2024;28(4):925\u0026ndash;37.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoganovic J. Familiarity with ChatGPT Features Modifies Expectations and Learning Outcomes of Dental Students. Int Dent J. 2024;74(6):1456\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaravia-Rojas M\u0026Aacute;, Camarena‐Fonseca AR, Le\u0026oacute;n‐Manco R, Geng‐Vivanco R. Artificial intelligence: ChatGPT as a disruptive didactic strategy in dental education. J Dent Educ. 2024;88(6):872\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchoenhof R, Schoenhof R, Blumenstock G, Lethaus B, Hoefert S. Synthetic, non-person related panoramic radiographs created by generative adversarial networks in research, clinical, and teaching applications. J Dent. 2024;146:105042.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStephan D, Bertsch A, Burwinkel M, Vinayahalingam S, Al-Nawas B, Kaemmerer P, Thiem DGE. AI in Dental Radiology-Improving the Efficiency of Reporting With ChatGPT: Comparative Study. J Med Internet Res. 2024;26(7):e60684.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTadinada A, Gul G, Godwin L, Al Sakka Y, Crain G, Stanford CM, Johnson J. Utilizing an organizational development framework as a road map for creating a technology-driven agile curriculum in predoctoral dental education. J Dent Educ. 2023;87(3):394\u0026ndash;400.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMahrous A, Botsko DL, Elgreatly A, Tsujimoto A, Qian F, Schneider GB. The use of artificial intelligence and game-based learning in removable partial denture design: A comparative study. J Dent Educ. 2023;87(8):1188\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYilmaz C, Erdem RZ, Uygun LA. Artificial intelligence knowledge, attitudes and application perspectives of undergraduate and specialty students of faculty of dentistry in Turkey: an online survey research. BMC Med Educ. 2024;24(1):1149\u0026ndash;1112.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmiri H, Peiravi S, rezazadeh shojaee Ss, Rouhparvarzamin M, Nateghi MN, Etemadi MH, ShojaeiBaghini M, Musaie F, Anvari MH, Asadi Anar M. Medical, dental, and nursing students\u0026rsquo; attitudes and knowledge towards artificial intelligence: a systematic review and meta-analysis. BMC Med Educ. 2024;24(1):412\u0026ndash;412.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnnamma LM, Varma SR, Abuttayem H, Prasad P, Azim SA, Odah R, George BT, Nair C, Karobari MI. Current challenges in dental education- a scoping review. BMC Med Educ. 2024;24(1):1523\u0026ndash;1510.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim CS, Samaniego CS, Sousa Melo SL, Brachvogel WA, Baskaran K, Rulli D. Artificial intelligence (A.I.) in dental curricula: Ethics and responsible integration. J Dent Educ. 2023;87(11):1570\u0026ndash;3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, Cates CJ, Cheng H-Y, Corbett MS, Eldridge SM, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCritical Appraisal Checklist for Analytical Cross Sectional Studies. [\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://jbi.global/sites/default/files/2020-08/Checklist_for_Analytical_Cross_Sectional_Studies.pdf]\u003c/span\u003e\u003cspan address=\"https://jbi.global/sites/default/files/2020-08/Checklist_for_Analytical_Cross_Sectional_Studies.pdf]\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHong QN, F\u0026agrave;bregues S, Bartlett G, Boardman F, Cargo M, Dagenais P, Gagnon M-P, Griffiths F, Nicolau B, O\u0026rsquo;Cathain A, et al. The Mixed Methods Appraisal Tool (MMAT) version 2018 for information professionals and researchers. Educ Inform. 2018;34(4):285\u0026ndash;91.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCritical Appraisal Checklist for Quasi-Experimental Studies. [\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://jbi.global/sites/default/files/2020-08/Checklist_for_Quasi-Experimental_Appraisal_Tool.pdf]\u003c/span\u003e\u003cspan address=\"https://jbi.global/sites/default/files/2020-08/Checklist_for_Quasi-Experimental_Appraisal_Tool.pdf]\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaethge C, Goldbeck-Wood S, Mertens S. SANRA\u0026mdash;a scale for the quality assessment of narrative review articles. Res Integr Peer Rev. 2019;4(1):5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJohn RM. AI ethics in dental practice. Br Dent J. 2024;236(9):698\u0026ndash;698.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWartman SA, Combs CD. Reimagining medical education in the age of AI. AMA J ethics. 2019;21(2):146\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRohlin M, Petersson K, Svens\u0026auml;ter G. The Malm\u0026ouml; model: a problem-based learning curriculum in undergraduate dental education. Eur J Dent Educ. 1998;2(3):103\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eObermeyer Z, Powers B, Vogeli C, Mullainathan S. Dissecting racial bias in an algorithm used to manage the health of populations. Science. 2019;366(6464):447\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRieke N, Hancox J, Li W, Milletari F, Roth HR, Albarqouni S, Bakas S, Galtier MN, Landman BA, Maier-Hein K. The future of digital health with federated learning. NPJ Digit Med. 2020;3(1):1\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuchanan C, Howitt ML, Wilson R, Booth RG, Risling T, Bamford M. Predicted influences of artificial intelligence on nursing education: Scoping review. JMIR Nurs. 2021;4(1):e23933.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eExploring how AI. can enhance dental education [\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.hsdm.harvard.edu/news/exploring-how-ai-can-enhance-dental-education]\u003c/span\u003e\u003cspan address=\"https://www.hsdm.harvard.edu/news/exploring-how-ai-can-enhance-dental-education]\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDentistry. Validation Dataset Guidance for Image Analysis Systems Using Artificial Intelligence [\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ada.org/-/media/Project/ADA%20Organization/ADA/ADA-org/Files/Resources/Practice/Dental%20Standards/ADA_1110-1_2025]\u003c/span\u003e\u003cspan address=\"https://www.ada.org/-/media/Project/ADA%20Organization/ADA/ADA-org/Files/Resources/Practice/Dental%20Standards/ADA_1110-1_2025]\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGDPR compliance checklist. [\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.onetrust.com/solutions/gdpr-compliance/]\u003c/span\u003e\u003cspan address=\"https://www.onetrust.com/solutions/gdpr-compliance/]\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDucret M, Wahal E, Gruson D, Amrani S, Richert R, Mouncif-Moungache M, Schwendicke F. Trustworthy Artificial Intelligence in Dentistry: Learnings from the EU AI Act. J Dent Res. 2024;103(11):1051\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDykstra S, MacDonald M, Beaudry R, Labib D, King M, Feng Y, Flewitt J, Bakal J, Lee B, Dean S, et al. An institutional framework to support ethical fair and equitable artificial intelligence augmented care. npj Digit Med. 2025;8(1):84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarte M, Carey B, Feng QJ, Alqarni A, Albuquerque R. Transforming undergraduate dental education: the impact of artificial intelligence. Br Dent J. 2025;238(1):57\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJobin A, Ienca M, Vayena E. The global landscape of AI ethics guidelines. Nat Mach Intell. 2019;1(9):389\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eADA technical report calls for external validation. of AI algorithms in dentistry [\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://adanews.ada.org/ada-news/2024/october/ada-technical-report-calls-for-external-validation-of-ai-algorithms-in-dentistry/]\u003c/span\u003e\u003cspan address=\"https://adanews.ada.org/ada-news/2024/october/ada-technical-report-calls-for-external-validation-of-ai-algorithms-in-dentistry/]\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePalaniappan K, Lin EYT, Vogel S. Global Regulatory Frameworks for the Use of Artificial Intelligence (AI) in the Healthcare Services Sector. Healthc (Basel) 2024, 12(5).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Artificial intelligence (AI), Dental education, Ethical challenges, Curriculum design, Clinical training, Accreditation standards, Algorithmic bias","lastPublishedDoi":"10.21203/rs.3.rs-6498214/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6498214/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e The integration of artificial intelligence (AI) into dental education offers transformative potential for enhancing learning outcomes, clinical training, and institutional efficiency. However, rapid AI adoption introduces ethical, logistical, and pedagogical challenges that require systematic exploration. This scoping review maps the current applications, challenges, and future directions of AI in dental education, focusing on its integration into curricula while ensuring ethical, equitable, and pedagogically sound practices.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e The Joanna Briggs Institute framework was followed, with reporting per the PRISMA-ScR guidelines for scoping reviews. A systematic search was conducted across PubMed, EMBASE, MEDLINE-Ovid, and Google Scholar for studies published between January 2018 and January 2025. The search terms included \"artificial intelligence,\" \"dental education,\" \"machine learning,\" \"ChatGPT,\" and \"ethical challenges,\" with Medical Subject Headings (MeSH) terms applied where applicable. After duplicate removal, 624 510 records underwent title/abstract screening, followed by a full-text review of 57 articles, with 43 studies meeting the eligibility criteria. Data extraction focused on the study design, population, AI type, key outcomes, and challenges.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e The key findings include the following: 1. AI-Driven Personalization: Generative AI (e.g., ChatGPT) reduced grading time by 45% and improved reflective learning outcomes, although 33% of studies reported algorithmic bias due to nonrepresentative training data. 2. In clinical training, AI tools achieved 99% accuracy in caries detection compared with 77–79% accuracy for students, but models trained on homogeneous datasets underperformed in diverse cohorts. 3. \u003cstrong\u003eInstitutional Efficiency\u003c/strong\u003e: Automated scheduling reduced administrative workloads by 30%, yet only 18% of institutions had updated curricula to include AI literacy modules. 4. Ethical Governance: Data privacy and data protection breaches occurred in 24% of the studies, and 41% reported faculty resistance to AI adoption, highlighting the need for dental-specific guidelines.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eAI holds significant promise for dental education but requires addressing ethical, logistical, and pedagogical challenges. Future efforts should focus on updating accreditation standards, fostering interdisciplinary collaboration, and developing hybrid models that balance AI-driven efficiency with traditional mentorship. Longitudinal studies are needed to evaluate the long-term impact of AI on clinical competence and patient outcomes.\u003c/p\u003e\n\u003cp\u003eSignificance: Dental educators need clearer guidance on integrating AI into the dental curriculum.\u003c/p\u003e","manuscriptTitle":"Artificial Intelligence in Dental Education: A Scoping Review of Opportunities, Challenges, and Ethical Frameworks for Shaping Accreditation Standards and Future Practice","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-05 09:40:14","doi":"10.21203/rs.3.rs-6498214/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"14f7f5c1-93bb-4e4d-8456-6fa0d0456ba6","owner":[],"postedDate":"May 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-06-06T04:08:32+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-05 09:40:14","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6498214","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6498214","identity":"rs-6498214","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}