Advancing Decision-Making through AI-Human Collaboration: A Systematic Review and Conceptual Framework

Advancing Decision-Making through AI-Human Collaboration: A Systematic Review and Conceptual Framework | 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 Advancing Decision-Making through AI-Human Collaboration: A Systematic Review and Conceptual Framework HAN LI, Feng Tian This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6885768/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 03 Apr, 2026 Read the published version in Group Decision and Negotiation → Version 1 posted You are reading this latest preprint version Abstract The interplay between humans and artificial intelligence (AI) in decision-making has become increasingly intricate and significant. Despite rapid advancements, the literature remains fragmented, with limited integrative frameworks to explain how AI-human dynamics and decision-making typologies shape outcomes. This study addresses this critical gap by conducting a systematic review and bibliometric analysis of 1,004 articles, culminating in a novel conceptual framework. The framework identifies two critical dimensions, AI-human dynamics and decision typologies, that shape decision outcomes and introduces four distinct paradigms of AI-human collaborative decision-making: autonomous execution, guided resolution, collaborative exploration, and augmented discovery. By synthesizing these paradigms, this research advances the theoretical understanding of hybrid decision-making systems and provides actionable insights for organizations navigating complex and AI-driven environments. By elucidating the mechanisms and trade-offs inherent in AI-human collaboration, this work lays a robust foundation for future research on adaptive decision systems in an era marked by accelerating technological change. Artificial Intelligence Decision-Making AI-Human Collaboration Figures Figure 1 Figure 2 Figure 3 Figure 4 INTRODUCTION Artificial intelligence (AI) is recognized as a transformative force reshaping employees, teams, and organizations (Cascio & Montealegre, 2016 ). AI decision-making is defined as the process of using artificial intelligence (AI) to support, enhance, or revolutionize the decision-making process, effectively incorporating AI technology into the decision-making process (Russell & Norvig, 2022 ; Brynjolfsson & McAfee, 2017 ). This new decision-making framework includes leveraging various AI technologies such as machine learning, natural language processing, and data analytics to process and analyze large volumes of data, identify patterns, and generate actionable insights (Goodfellow, Bengio, & Courville, 2016 ; McKinsey Global Institute, 2018 ; Ng, 2016 ; Kim, Lee, & Hwang, 2021 ; Lee et al., 2023 ). Organizations are increasingly adopting AI to enhance decision quality, driven by demands for speed, accuracy, and adaptability in volatile environments (Akhtar et al., 2019 ; Vrontis et al., 2022 ). For example, AI applications range from machine learning algorithms predicting market trends (Fountaine et al., 2019 ) and AI-driven analytics optimizing supply chain operations (Choudhury et al., 2020 ), to AI-based tools supporting financial decision-making (Davenport & Ronanki, 2018 ). A study conducted by Oracle and Future Workplace ( 2019 ) revealed a growing influence of AI, with 50 percent of employees presently applying various forms of AI to assist in decision-making. Despite these significant advantages, integrating AI into decision-making processes not only promises to boost efficiency and effectiveness but also introduces new complexities and challenges that require a deeper understanding of its mechanisms and outcomes (Choudhury et al., 2020 ). The rapid adoption of AI tools has not been as smooth as anticipated (Arslan et al., 2022 ), and the technological inflexibility potentially may lead to organizational disruptions, thereby diminishing the advantages of AI in decision-making and escalating governance costs (Basu et al., 2023 ). Given the extensive applications of AI in decision-making, the emerging research field of AI-empowered decision-making has garnered more attention from both scholars and practitioners. To promote academic development in this field, some researchers have attempted to review or analyze relevant literature to guide future research. For example, Borges (2021) identified key sources of value creation, such as decision support, customer engagement, automation, and the development of new products and services, emphasizing AI's potential to transform organizational strategy. Pietronudo (2021) argues that while AI’s ability to rationalize decision-making and foster creativity has been acknowledged, its impact on organizational processes remains multifaceted, leading to both expected and unexpected outcomes. At the same time, scholars have proposed hybrid models that combine human intelligence with AI to address complex problems. These models suggest that AI can expand the range of potential solutions, thus complementing human decision-making (Raisch & Fomina, 2023 ). A framework for integrating human and AI decision-making emphasizes key factors such as search space specificity, decision speed, and transparency, while also addressing challenges like AI bias and the need for accountability in algorithmic decision-making (Shrestha et al., 2019 ). Additionally, research on AI's impact on the workforce reveals its significant influence on job roles, required skills, and employee attitudes. As AI continues to reshape organizational practices and labor markets, understanding its long-term effects on innovation and workforce dynamics is crucial (Bankins et al., 2024 ). While prior studies explore narrow aspects of AI’s role, such as automation (Brynjolfsson & McAfee, 2017 ), decision support (Shrestha et al., 2019 ), or ethical implications (Raisch & Krakowski, 2021 ), they lack a holistic perspective. Existing reviews either focus on technical capabilities, human-AI interaction models, or organizational impacts in isolation, leaving a fragmented understanding of how these dimensions interact (Raisch & Fomina, 2023 ; Bankins et al., 2024 ). This fragmentation obscures the mechanisms through which AI reshapes decision processes and inhibits the development of actionable frameworks for practitioners. This study addresses two critical gaps in the existing literature. Theoretically, prior research has suffered from fragmentation due to the absence of integrative frameworks that explain how AI-human collaboration varies across different decision contexts, such as structured versus unstructured problems, and power dynamics, including AI-led versus human-led processes. While several systematic reviews have tackled decision-making in complex contexts such as multiobjective optimization (Pajasmaa et al., 2025 ), few have specifically focused on the evolving dynamics of AI-human collaboration across decision typologies. Practically, organizations face challenges in aligning AI adoption with decision-making needs, often overemphasizing automation at the expense of creativity and ethical considerations. To bridge these gaps, we conducted a systematic review and bibliometric analysis of 1,005 articles, which culminated in the development of a novel AI-Human Decision Matrix. This framework categorizes decision-making into four paradigms, autonomous execution, guided resolution, collaborative exploration, and augmented discovery, based on two dimensions: AI-human dynamics (ranging from dominance to empowerment) and decision typology (spanning open to closed). By mapping these paradigms to various organizational contexts, our study not only provides a strategic tool for optimizing AI adoption but also offers meaningful insights for addressing ethical and operational challenges in the rapidly evolving landscape of decision-making. To structure our inquiry, we propose the following research questions: (1) How have scholars conceptualized and classified the roles of AI in organizational decision-making across various contexts? (2) What are the dominant patterns, paradigms, and trends in AI-human collaborative decision-making revealed through bibliometric analysis? (3) How can we develop an integrative framework that maps AI-human dynamics against different decision typologies to inform both academic inquiry and practical application? By addressing these questions, this study aims to synthesize fragmented insights into a coherent framework, deepening our understanding of hybrid decision-making systems and informing responsible AI deployment in complex organizational environments. METHODOLOGY Data Collection Literature data is sourced from the Web of Science (WoS) Core Collection, concentrating on the Social Sciences Citation Index (SSCI) and the Science Citation Index Expanded (SCI-EXPANDED). As highlighted in prior studies (Jiang et al.,2024; Pouris & Pouris, 2011; Mongeon & Paul-Hus, 2016; Borges, 2021; Adriaanse & Rensleigh, 2013), the Web of Science (WoS) database serves as the primary source for data collection. The rationale for selecting the WoS database is multifaceted: First, it encompasses the most significant and influential journals in the field (Pouris & Pouris, 2011). Second, given the interdisciplinary nature of AI Decision-making, which includes information technology and strategic management, WoS is particularly well-suited for our objectives due to its comprehensive coverage across diverse disciplines (Adriaanse & Rensleigh, 2013). Third, WoS is recognized as one of the most prestigious and respected repositories of scientific information in the social sciences (Mongeon & Paul-Hus, 2016), and its utility has been extensively demonstrated in prior review articles related to management decision-making (Borges, 2021). The screening process is illustrated in Fig. 1 . In this study, we focused on journal articles published between 2010 and 2024. The academic literature in the field of AI and Decision-making began in the 1990s, but articles related to AI applications became prominent only after 2010. Therefore, we selected this time frame from 2010 to 2024 to align with the notable growth in AI and machine learning applications in management. The primary search terms were constructed using Boolean operators (AND/OR) to link two groups of keywords. The first group included terms such as “decision quality,” “decision efficiency,” “decision effectiveness,” or simply “decision.” The second group contained keywords referencing AI-related technologies, such as “artificial intelligence,” “machine learning,” or “AI.” The search was restricted to English-language journal articles, as these are widely regarded as representative of the most influential and reputable research. This process resulted in a total of 28,492 articles. Given the multidisciplinary nature of AI research, we refined the results to ensure relevance to management studies. Specifically, we limited the Web of Science categories to include only publications related to management, business, economics, and business finance. This refinement resulted in a total of 1,005 articles. Data Analysis The bibliographic records were exported from the Web of Science (WoS) database to VOSviewer 1.6.19 for bibliometric analysis. Compared to other bibliometric tools, VOSviewer offers better visualization capabilities (Van Eck & Waltman, 2010). VOSviewer focuses on the knowledge structure of specific domains, enabling researchers to identify and visualize hidden patterns and trends by extracting information from a large number of bibliographic records (Cobo et al., 2011). Each record includes details such as authors and their affiliations including countries, titles, abstracts, keywords, publication years, source journals, and references. Based on the bibliographic data, analyses of keyword co-occurrence, trend dynamics, and country-wise literature contributions were conducted. Keyword co-occurrence A keyword co-occurrence analysis was conducted to identify relationships among themes within the literature. The co-occurrence network visualization map indicated that these clusters formed a broad and interconnected topic network within the studies (Gupta & Khan, 2024). The software detected links between similar keywords and grouped them into clusters, each represented by a distinct color. Keywords within the same cluster reflect similar or closely related themes (Ali & Gölgeci, 2019; van Eck & Waltman, 2009). The results of this analysis are presented in Fig. 2 . The existing themes in AI-related decision-making literature were categorized into four clusters. The red cluster (N = 10 terms/keywords) focuses on automated decision-making, comprising research on decision-making. The green cluster (N = 11 terms/keywords) is labeled as augmentation and collaboration, comprising research on decision-making. The blue cluster (N = 15 terms/keywords) is identified as strategic discovery and innovation. The yellow cluster (N = 10 terms/keywords) highlights aspects such as predictive analytics and data analytics. The initial red-colored cluster (N = 10 terms/keywords) is “Autonomous Execution” consists of "Decision-making", "Automation", "Adoption" and "Trust", which are frequently repeated keywords. In the yellow cluster (N = 10 terms/keywords), Guided Resolution, the most frequently repeated keywords are "Efficiency", " Predictive Analytics" and " Data Analytics ". The third cluster in green (N = 11 terms/keywords) is Collaborative Exploration (AI-dominated Open Decision-Making), composed of the most frequently repeated keywords: "Augmentation", "Creativity", "Rational" and "Decision support systems". Lastly, "Artificial Intelligence (AI)", "Innovation", and "Complexity" are the most frequently repeated keywords in the blue cluster (N = 15 terms/keywords) Augmented Discovery. Authorship analysis Table 1 highlights the leading contributors in articles related to AI decision management, focusing on their publication volume and citation impact. Among the 226 authors analyzed, the ten most highly cited authors are presented, with nine of them receiving over 200 citations. Table 2 demonstrates that Shivam Gupta stands out as the most prolific author, with 10 publications and 457 citations. He is followed by Yogesh Dwivedi, who, despite having published fewer papers (7), achieved the highest citation count of 544. Other notable contributors include Patrick Mikalef (6 publications, 265 citations), Mehrbakhsh Nilashi (5 publications, 281 citations), and Ajay Kumar (5 publications, 274 citations). The remaining authors each have four publications, with Vinit Parida leading this group with 338 citations, followed by Yash Raj Shrestha, Dimitris Bertsimas, Krogh Georg Von, and Arpan Kumar Kar, who demonstrate varying degrees of citation impact. Table 1 Top-cited authors for bibliometric reviews Author Papers Papers citations Yogesh Dwivedi 7 544 Shivam Gupta 10 457 Vinit Parida 4 338 Yash Raj Shrestha 4 322 Dimitris Bertsimas 4 307 Krogh Georg Von 4 304 Mehrbakhsh Nilashi 5 281 Ajay Kumar 5 274 Patrick Mikalef 6 265 Arpan Kumar Kar 4 193 In addition to citation performance analysis, we employed a co-authorship network mapping to identify the primary groups of contributors to AI decision management literature. The co-authorship network encompasses 21 authors who have each published at least four bibliometric review articles and received at least one citation. This analysis reveals two major clusters, as illustrated in Fig. 6, with Gupta Shivam emerging as the central figure in terms of both publication count and citation influence. RESULTS To provide an overview of existing research on AI decision-making, we use bibliometric analysis to conduct a comprehensive literature review. Artificial intelligence significantly enhances the speed and effectiveness of decision-making by assisting humans in selecting among various options and generating different alternatives for human choice. Various types of AI decision-making are applied in different social and business contexts. This demonstrates the complementary relationship between humans and AI in the decision-making process (Von Krogh, 2018). Through the results of bibliometric analysis, after synthesizing the four clusters presented in Fig. 4 , we identified and classified an AI decision matrix with the horizontal axis labeled “Human-AI Dynamics” and the vertical axis labeled “Decision-making Paradigm.” This matrix allowed us to recognize four distinct AI decision-making types: "Autonomous Execution," which refers to AI-dominated closed decision-making where the system operates independently to reach a conclusion; "Guided Resolution," which refers to AI-empowered closed decision-making where AI supports but does not fully control the decision process; "Collaborative Exploration," which refers to AI-dominated open decision-making where AI leads the exploration and decision-making process in an open-ended context; and "Augmented Discovery," which refers to AI-empowered open decision-making where AI assists human decision-makers in exploring and discovering new possibilities in uncertain environments. We defined four types of AI decision-making by using the horizontal axis to represent AI-dominated versus human-dominated decisions, and the vertical axis to represent open versus closed decision-making. The difference among the four AI decision types is based on whether the final decision is made by humans with the support of AI or made by AI with humans’ supervision, and whether the decision involves a well-defined, structured problem or a diverse, flexible problem with multiple possible answers. By evaluating the effectiveness and applicability of different decision-making models, we can identify scenarios where AI-dominated decisions are more suitable or when human judgment and feedback are necessary. Consequently, we propose four types of AI decision-making: "Autonomous Execution" (AI-dominated closed decision-making), "Guided Resolution" (AI-empowered closed decision-making), "Collaborative Exploration" (AI-dominated open decision-making), and "Augmented Discovery" (AI-empowered open decision-making). These types aid in understanding the roles and impacts of different AI decision-making models. The largest cluster represents the "Autonomous Execution" type of AI decision-making, characterized by AI-dominated closed decision-making. The core of this decision-making model lies in delegating certain tasks or operations based on closed decision-making to machines, enabling more comprehensive, rational, and efficient processing. The second largest cluster embodies the "Guided Resolution" type of AI decision-making, where AI plays a supportive and enabling role in the decision-making process rather than making decisions autonomously. The third largest cluster represents the "Collaborative Exploration" model of AI decision-making. In this model, AI leads the process through alternative decision-making, assuming responsibility for the final decision, while humans provide support and feedback to help AI optimize and adjust decisions. The last cluster represents the "Augmented Discovery" type of AI decision-making, focusing on leveraging AI to enhance human capabilities, thereby driving discovery and innovation in open decision-making. Building upon prior literature, we categorize the decision-making process of each type into three distinct stages, problem identification, information collection and processing, and decision implementation (Simon, 1960; March, 1994). These stages serve as a framework to analyze and distinguish the contributions of each study within the clusters, enhancing our ability to relate them to broader decision-making dynamics. Within the decision-making process, individuals must navigate complex environments to acquire critical decision-related information, analyze contextual factors and potential outcomes, and determine whether favorable conditions for action exist (Klein, 1998; Gigerenzer & Goldstein, 1996). Significant variation exists among individuals in their understanding and practice of decision-making, driven by factors such as cognitive capabilities (Kahneman, 2011), accumulated experience (Argyris & Schön, 1978), organizational support (Mintzberg et al., 1976), and external environmental uncertainty (Tversky & Kahneman, 1974). The final stage of decision-making is implementation which becomes particularly critical in high-risk and complex scenarios, where effective resource allocation and team collaboration are essential to increase the likelihood of success (Cyert & March, 1963; Bazerman & Moore, 2013). Fully committing to the execution phase fosters the attainment of organizational or individual objectives, enabling broader progress (Gollwitzer, 1999; Heath & Heath, 2010). The different types of AI decision-making exhibit distinct characteristics across the three stages of information collection, information processing, and decision implementation. Below, we describe the features and differences of the four decision-making types. Table 2 summarizes the four clusters and decision types. Table 2 Clusters and Decision Types Decision Types Autonomous Execution (Red) Guided Resolution (Green) Collaborative Exploration (Yellow) Augmented Discovery (Blue) Keywords Automation; Adoption; Decision making; Determinants; Acceptance. Augmentation; Rational; Collaboration; Uncertainty; Decision support systems. Efficiency; Prediction; Simulation; Strategic judgment. Innovation; Strategy; Complexity; Discovery; Implementation. Definition AI-dominated closed decision-making where the system operates independently to reach a conclusion. AI-empowered closed decision-making where AI supports but does not fully control the decision process AI-dominated open decision-making where AI leads the exploration and decision-making process in an open-ended context. AI-empowered open decision-making where AI assists human decision-makers in exploring and discovering new possibilities in uncertain environments Information Collection AI leads the entire data collection process by automating the real-time gathering, processing, and integration of data from various structured sources, ensuring high quality and consistency through techniques like data fusion, monitoring, and cleaning. AI significantly enhances human decision-making by efficiently processing and analyzing large datasets across various sectors, such as healthcare and finance, thereby extending cognitive capabilities and improving the accuracy and speed of complex analyses. AI synthesizes data from diverse, unstructured sources to detect trends and shifts, while human input ensures adaptive strategy adjustments, exemplifying the importance of cross-domain data fusion and real-time analytic in gaining competitive advantages. AI empowers humans by rapidly analyzing complex and unstructured data to uncover latent patterns and emerging trends, significantly enhancing the comprehensiveness, precision, and efficiency of data collection, as demonstrated in social sciences research. Information Process AI uses advanced techniques to analyze large datasets, identify patterns, make predictions and generate actionable insights. AI empowers data-driven decision-making by providing analytical insights through its computational power, while humans maintain interpretative control, exemplified in business management and supply chain optimization where AI's analysis supports strategic human judgment. AI utilizes advanced methodologies such as machine learning, deep learning, and reinforcement learning to uncover latent patterns and create predictive models, enabling actionable forecasts and innovative solutions to open-ended challenges, as demonstrated by Netflix's content strategy optimization. The information processing phase involves intensive collaboration between AI and humans, where AI's predictive modeling is refined by human expertise and creativity, integrating emotional intelligence with data-driven decision-making to foster innovation, as seen in personalized medical treatment planning. Decision Implementation AI not only processes information but also autonomously makes and executes decisions with minimal human involvement, ensuring rapid, real-time adjustments and optimizing processes. AI offers data-driven support, while humans integrate these insights with their expertise and situational judgment, balancing rational and emotional considerations, as seen in healthcare and emergency management contexts. AI autonomously generates actionable recommendations and leads decision execution, while human oversight ensures iterative refinements, exemplified in new product development where AI designs prototypes and humans validate and adjust outcomes. The decision implementation phase epitomizes human-AI collaboration, where humans integrate AI-generated analyses with intuition, experience, and value-based judgments to execute decisions, exemplified in artistic creation where AI provides inspiration and artists refine outputs to reflect personal style and emotion, showcasing AI-empowered open decision-making. Cluster 1: Autonomous Execution In the realm of Autonomous Execution, the role of artificial intelligence is to orchestrate every facet of information collection, processing, and decision implementation. Autonomous Execution can be conceptualized as an AI-dominated closed decision-making process. Within Autonomous Execution, artificial intelligence is tasked not only with acquiring and integrating information from diverse sources but also with conducting in-depth analysis and processing of this information to formulate optimal decisions. At last, AI autonomously implements these decisions based on the analytical outcomes, thereby achieving efficient and precise operations. This comprehensive leadership capability positions artificial intelligence as a pivotal element in Autonomous Execution, ensuring the seamless and effective progression of the entire process. Information Collection Stage In the Autonomous Execution model, the information collection process is entirely led by artificial intelligence (AI). This process typically relies on data from multiple standardized and structured sources, such as sensors, operation logs, enterprise resource planning (ERP) systems, and Internet of Things (IoT) devices(Rane & Narvel, 2022 ). AI automates the real-time collection and processing of this data to ensure high quality and consistency. Specifically, the data collection process often involves multiple technical methods, including data transmission, real-time monitoring, and data cleaning. By utilizing technologies such as Natural Language Processing (NLP), sensor data fusion, and rule-based reasoning engines, AI integrates large volumes of dispersed data to ensure its accuracy and reliability (Balasubramanian et al., 2022 ). For instance, in intelligent manufacturing systems, sensors collect real-time data on equipment parameters such as temperature, vibration, and pressure. AI uses this data to monitor equipment performance and ensure the smooth operation of production processes༈Davenport & Ronanki, 2018 ). In this process, AI not only executes data collection but also plays a central role in ensuring the effectiveness of the data, preventing noise or anomalies from compromising the accuracy of decision-making. Information Processing Stage During the information processing stage, it can highlight the core advantages of AI algorithms, including the rapid identification of patterns and trends, as well as the efficient processing of large-scale datasets. AI performs deep processing on the massive data collected to extract useful patterns, trends, or anomalies. AI employs various advanced techniques, including machine learning, deep learning, reinforcement learning, and data mining, to train algorithms and analyze data (Chatterjee & Vrontis, 2022). These techniques help AI identify underlying patterns within the data, make predictions, and recognize patterns. For example, on a production line, AI can use historical data to train models that predict equipment failure times and provide maintenance recommendations (Serradilla et al., 2022 ). In the information processing stage, it is characterized by AI's ability to efficiently process data and generate useful information for decision-making through model-based algorithms and data analysis, reflecting the characteristics of rational decision-making (Simon, 1979). For example, in autonomous driving, AI processes data from radar and cameras and uses deep neural network models to classify and analyze images, recognize traffic signs, pedestrians, and other vehicles, and predict future driving conditions. These processed results form the basis for AI's decisions, enabling it to respond effectively in complex traffic environments. AI optimizes the decision-making process through algorithms, avoids human intervention, and improves system decisions' efficiency and accuracy (Shrestha et al., 2019 ). Decision Implementation Stage In the Autonomous Execution practice, AI not only processes information but also fully takes on the responsibility of executing decisions, with minimal human involvement. In this model, AI automatically implements decisions based on the results of prior information collection and processing, adjusting in real-time according to feedback. For example, in industrial manufacturing, robots automatically adjust machine operations based on AI-generated instructions, optimizing the production process (Grigorievich, 2024 ). Autonomous vehicles execute braking or steering decisions made by AI to avoid accidents (Bathla et al., 2022 ). At this stage, AI ensures the rapid and efficient execution of decisions through its precise execution capabilities. In this process, AI is not only the decision-maker but also the executor of those decisions. The advantage lies in AI's ability to respond instantly, with execution speeds far exceeding human capabilities. Additionally, AI can adjust based on real-time feedback, ensuring precision and consistency throughout the execution process. The decision-execution phase emphasizes minimizing human intervention, reducing human error, and improving the efficiency and accuracy of decision-making and execution (Brynjolfsson & McAfee, 2023 ). Autonomous Execution further demonstrates that AI systems can address decision-making problems in closed environments, particularly in applications requiring high precision and real-time capabilities. Through information control and execution accuracy, AI maximizes system efficiency. For example, AI systems in fully automated factories can precisely control every detail of the production line, ensuring continuous production and product quality (Sahoo & Lo, 2022 ). Cluster 2: Guided Resolution In the Guided Resolution model, artificial intelligence is strategically positioned as a tool and support system for human decision-making, with the primary objective of augmenting human capabilities rather than supplanting them entirely. Guided Resolution can be understood as AI-empowered closed decision-making. Within this framework, artificial intelligence assists by providing functionalities such as data analysis, pattern recognition, and predictive insights, thereby enabling humans to make more effective judgments and actions within closed decision-making environments. The involvement of AI allows humans to better comprehend complex information and contexts, thus enhancing the accuracy and efficiency of decision-making. Through this collaborative approach, artificial intelligence not only elevates human decision-making capabilities but also ensures that humans maintain a leading role throughout the process, fully leveraging their creativity and judgment. This model of human-machine collaboration underscores the value of artificial intelligence as an auxiliary tool, facilitating the maximization of human potential. Information Collection Stage During the information collection stage, AI aids humans in processing vast amounts of data by filtering, categorizing, and optimizing information, thus providing a clearer foundation for decision-making. For instance, in healthcare, AI supports physicians by analyzing patient health data, such as medical images, genetic test results, or electronic health records, to quickly extract key indicators, thereby accelerating the diagnostic process. This model extends human cognitive and analytical capabilities, demonstrating significant advantages in complex, high-precision environments (Topol, 2019; Daugherty & Wilson, 2018). In the financial sector, AI analyzes market trends and risk factors. Natural language processing (NLP) techniques enable AI to rapidly filter financial news for investment-relevant information, optimizing portfolio strategies (Chui et al., 2018). These tools enhance the efficiency of information collection and ensure the reliability of analysis results through automated data cleaning and standardization (Marr, 2021). Information Processing Stage During information processing, AI leverages its computational power to generate analytical results for human reference. While AI plays a dominant role in analysis, humans retain control over interpreting and utilizing the outcomes. For instance, in business management, AI can analyze customer data to generate recommendations for marketing strategies, such as identifying the most effective advertising times and target audiences (Goodfellow, Bengio, & Courville, 2016 ). This process reflects the core principle of Data-Driven Decision-Making, emphasizing the use of data analytics to provide scientific support for decision-making (Brynjolfsson & McAfee, 2014). In supply chain management, AI employs optimization algorithms and predictive models to help businesses accurately estimate inventory needs and optimize logistics routes, reducing operational costs. This model combines AI's analytical capabilities with human strategic judgment. For example, under conditions of high market uncertainty, humans may adjust production plans based on AI's forecasts to mitigate potential risks (Agrawal et al., 2018; McKinsey Global Institute, 2020). Decision Implementation Stage In the final stage of decision-making and implementation, AI provides auxiliary support, while humans make decisions by integrating AI-generated insights with their own knowledge and situational judgment. For example, in healthcare, physicians may use AI-generated treatment plans as a reference but adapt them during implementation based on patient conditions, personal preferences, and ethical considerations. This approach integrates Rational Decision-Making with Emotional Decision-Making, where AI provides data-driven support, while humans evaluate the appropriateness of decisions using ethics and professional expertise (Lerner et al., 2015; Topol, 2019). In emergency management, AI can analyze real-time data to suggest optimal response strategies, such as resource allocation during natural disasters. However, the final action plan requires human deliberation, considering social factors, political decisions, and evolving on-the-ground circumstances (Brynjolfsson et al., 2017; Edmondson & Harvey, 2018). Cluster 3: Collaborative Exploration In the Collaborative Exploration process, artificial intelligence assumes a leading role in addressing open-ended and highly complex decision-making problems. Collaborative Exploration can be understood as AI-dominated open decision-making. Its defining feature lies in its capacity to conduct exploratory analyses to uncover unknown patterns or generate innovative solutions. Within this model, artificial intelligence aids in identifying potential opportunities and challenges by processing vast amounts of data and performing complex predictive modeling. This approach is particularly well-suited for contexts requiring extensive data processing and intricate predictive modeling, such as product development, new market exploration, or hypothesis generation in scientific research (Kaplan & Haenlein, 2023 ; Fleming, 2018). In these domains, artificial intelligence not only rapidly analyzes and integrates data from multiple sources but also identifies hidden trends and relationships, offering unprecedented insights and strategies. Through this mechanism, artificial intelligence enhances the innovativeness and effectiveness of decision-making within the Collaborative Exploration model, thereby advancing organizational adaptability and competitive advantage under conditions of uncertainty and complexity. Information Collection Stage During the information collection stage, AI extracts data from diverse and unstructured sources to construct a comprehensive informational foundation. By integrating data across domains, AI can quickly detect market trends and shifts. For example, natural language processing techniques are employed to analyze social media content, providing insights into consumer behavior(Davenport & Guha, 2020). Meanwhile, human input remains vital during this phase, enabling adaptive adjustments to data strategies in response to evolving demands (Chesbrough, 2003). In drug discovery, for instance, AI can synthesize genomic data, experimental findings, and academic literature to identify potential drug targets and therapeutic pathways (Fleming, 2018). This phase emphasizes the importance of cross-domain data fusion and real-time analytics, which collectively enable organizations to build competitive advantages in complex environments. Information Processing Stage AI employs advanced methodologies, including machine learning, deep learning, and reinforcement learning, to extract deeper insights and construct predictive models (Chatterjee & Vrontis, 2022). The goal is to identify latent patterns and generate actionable forecasts. For instance, in Netflix’s content strategy, AI analyzes user viewing preferences, historical consumption data, and cultural trends to forecast the most promising combinations of scripts, directors, and actors, thereby optimizing resource allocation for content production (Westcott Grant, 2018). This exploratory analytical capability allows AI to generate novel insights and provide solutions to open-ended challenges. Decision Implementation Stage In the implementation phase, AI takes on a proactive role by generating actionable recommendations and often leading the execution of final decisions. Nonetheless, human oversight and iterative refinement remain essential to ensure contextual relevance, ethical alignment, and adaptive flexibility. For instance, in new product development, AI can autonomously generate prototype designs aligned with customer preferences or offer conceptual blueprints that serve as a foundation for refinement by R&D teams. Within this decision-making configuration, AI drives the exploratory and generative tasks, while humans focus on validating, interpreting, and tailoring outputs to organizational objectives (Kaplan & Haenlein, 2023 ). Emerging studies further underscore the significance of mutual learning between human and AI agents in such settings, revealing how co-learning dynamics not only enhance outcome quality but also shape the evolving structure of collaboration within AI-augmented teams (Lu et al., 2025 ). Cluster 4: Augmented Discovery Augmented Discovery emphasizes the profound collaboration between artificial intelligence and humans within open decision-making contexts. Augmented Discovery can be understood as AI-empowered open decision-making. In this model, artificial intelligence, through its formidable computational power and data analytical support, assists humans in addressing complex problems while preserving the central role of human creativity, ethical considerations, and value judgment. The involvement of AI enables humans to process vast amounts of information and complex variables more effectively, thereby playing a crucial role in domains such as scientific research, artistic creation, and the resolution of intricate societal challenges (Topol, 2019; Lerner et al., 2015; Deranty & Corbin, 2024). In these areas, artificial intelligence not only provides technical support but also aids in identifying potential patterns and trends, helping humans explore new possibilities and solutions. Through this collaborative approach, the Augmented Discovery model enhances the depth and breadth of decision-making while ensuring human agency and responsibility in the face of uncertainty and complexity. This model of human-machine collaboration fosters innovation and societal progress while maintaining ethical and value-oriented human guidance in the decision-making process. Information Collection Stage During the information collection stage, AI aids humans in extracting insights from diverse data sources. The primary objective of this phase is to explore uncharted territories and uncover latent patterns. By analyzing complex and unstructured data, such as texts and images, AI helps identify emerging trends. In the field of social sciences, for instance, AI can analyze vast amounts of social media content to reveal intricate patterns of social behavior or the impacts of public policies (Deranty & Corbin, 2024). This phase exemplifies the empowering role of AI in data collection, wherein its rapid analytical capabilities significantly enhance the comprehensiveness, precision, and efficiency of the process. Information Processing Stage The information processing stage exemplifies a deeply synergistic interaction between artificial intelligence and human judgment. In this phase, AI employs predictive modeling techniques to produce preliminary insights, which are then enriched and contextualized by human experts drawing on their domain-specific knowledge and creative reasoning. This co-active process blends emotional intelligence, human creativity, and algorithmic precision, thereby enabling data-driven decisions that are both analytically robust and contextually nuanced (Shrestha et al., 2019 ). Recent research further highlights the role of natural language processing in supporting complex knowledge synthesis within collaborative environments, enhancing the efficiency and integration of decision-relevant insights in AI-supported settings (Gimpel et al., 2024 ). For instance, in healthcare, AI systems analyze genomic profiles and patient records to propose individualized treatment options, which physicians then refine by considering clinical realities and patient values to arrive at the final therapeutic decision (Topol, 2019). Decision Implementation Stage The decision implementation phase represents the pinnacle of human-AI collaboration. In this stage, humans synthesize AI-generated analyses with their intuition, experience, and value-based judgments to finalize and execute decisions. In the realm of artistic creation, for instance, AI can provide artists with creative inspiration by generating drafts of abstract paintings, which are further refined by artists to reflect their personal style and emotional expression. This process highlights the fusion of creativity and emotion, demonstrating the advantages of AI-empowered open decision-making (Lerner et al., 2015). DISCUSSION This study utilizes bibliometric analysis to systematically review the impact of AI on decision-making processes. By leveraging bibliometric methods, key themes and trends within the established academic literature are identified and analyzed.. The analysis has identified four critical themes that characterize AI's role in decision-making: Autonomous Execution, Guided Resolution, Collaborative Exploration, and Augmented Discovery. Building on these themes, a novel framework is proposed, categorizing AI's roles as either enabling or leading within structured (closed) and unstructured (open) decision-making environments.. The integration of AI into decision-making has given rise to a variety of models, each characterized by distinct strengths and limitations. Autonomous Execution demonstrates AI's ability to lead the entire decision-making process, encompassing data collection, analysis, and execution. It is particularly effective in closed decision environments, where AI minimizes human involvement, delivering efficient and accurate decisions while enhancing system automation and intelligence. This approach is widely applicable across industries requiring consistency and speed, such as manufacturing, logistics, and predictive maintenance. In contrast to Autonomous Execution, Guided Resolution focuses on collaboration, leveraging AI's computational and analytical strengths to support human decision-makers while retaining human authority.. This complementary approach is particularly beneficial in domains such as healthcare, finance, supply chain management, and emergency response. By integrating AI's precision with human intuition and ethical reasoning, the Guided Resolution model addresses complex, closed decision scenarios that require a balance of data-driven insights and contextual judgment. Collaborative Exploration emphasizes AI's dominant role in open-ended decision-making processes, leveraging its ability to integrate vast datasets and generate innovative solutions. Its application is particularly relevant in scenarios requiring exploration and adaptability, such as scientific research, product innovation, and strategic planning. However, AI's limitations in addressing uncertainties and value-based judgments highlight the need for human involvement in later stages to validate and refine AI-generated insights (Brynjolfsson & McAfee, 2014; Raisch & Krakowski, 2021 ). Augmented Discovery highlights a co-creative process where AI complements human creativity and expertise, making it ideal for open, dynamic environments. For example, during global health crises, AI can rapidly analyze transmission models and propose mitigation strategies, while human policymakers evaluate these recommendations through ethical and societal lenses (Davenport & Kirby, 2016). Despite its adaptability, challenges such as over-reliance on AI-generated suggestions and concerns about algorithmic transparency and ethics remain critical considerations (Raisch & Krakowski, 2021 ). Building on these observations, our study yields important implications for both theory and practice. Theoretical Implications This study makes several contributions to the theoretical understanding of AI decision-making. First, by proposing the AI-Human Decision Matrix, we categorize AI’s roles in decision-making as either enabling or leading, applied across both structured (closed) and unstructured (open) decision contexts. This framework synthesizes previously fragmented literature and clarifies how AI can complement or substitute human involvement at different stages of decision-making (Akhtar et al., 2019 ; Raisch & Krakowski, 2021 ). Second, the framework advances classical decision-making theories (e.g., Simon, 1960; March, 1994) by introducing a hybrid lens through which to understand AI’s impact on problem identification, information processing, and execution. It extends prior conceptualizations by distinguishing between AI-empowered and AI-dominated roles and mapping these onto specific decision paradigms such as Collaborative Exploration or Augmented Discovery. Third, our study opens new avenues for theorizing the cognitive division of labor between humans and machines. It reframes AI not merely as a tool for automation or support but also as a potential co-creator of strategic insight in open decision environments (Fleming, 2018; Brynjolfsson & McAfee, 2014). Finally, the findings underscore critical unresolved theoretical issues that merit further exploration: the need for interdisciplinary integration, longitudinal studies, contextualized analyses, and more nuanced models of human-AI collaboration, particularly as they relate to ethical norms and organizational governance (Arslan et al., 2022 ; Raisch & Krakowski, 2021 ). Practical Implications From a managerial perspective, the framework offers a practical roadmap for aligning AI deployment with organizational decision complexity. First, it enables organizations to assess the fit between AI capabilities and decision typologies. By understanding when to apply AI-led automation versus AI-empowered augmentation, managers can optimize efficiency while maintaining human oversight and creativity. This is particularly important given the trade-offs each model entails, for instance, between speed and transparency, or precision and adaptability. Second, the framework informs the design of governance mechanisms that distinguish between decision contexts requiring AI leadership and necessitating human control. Such differentiation is essential for managing risk, ensuring accountability, and fostering trust in AI-augmented systems (Davenport & Ronanki, 2018 ; Kaplan & Haenlein, 2023 ). Third, the matrix has implications for capability development. As AI becomes more deeply embedded in strategic and operational decision-making, organizations must equip their workforce with not just data literacy but also collaborative fluency, the ability to interact effectively with AI tools in both closed and open problem spaces. AI integration requires a balanced approach that enhances decision performance without compromising values such as fairness, transparency, and accountability (Brynjolfsson & McAfee, 2014; Fleming, 2018).. Future Research Led by the insights generated through our framework and literature analysis, we identify three key directions for future research. First, team-level AI-enabled decision-making remains an underexplored domain. While existing research has predominantly centered on individual or organizational decision contexts, there is a pressing need to investigate how AI integration influences group dynamics, including trust formation, coordination mechanisms, and collective performance (Cascio & Montealegre, 2016 ; Carmeli et al., 2019 ). In this regard, recent empirical work by Lu et al. ( 2025 ) highlights the pivotal role of human-AI co-learning in shaping team cognition and behavioral adaptation, offering a promising foundation for future inquiry into hybrid team functioning. Particular emphasis should be placed on understanding how AI affects psychological safety, role allocation, and the development of shared mental models within decision-making teams. Second, longitudinal and cross-contextual studies are needed to understand how AI-human collaboration evolves over time and across industries. These studies could assess the stability of different AI-human decision paradigms, such as Autonomous Execution versus Augmented Discovery, and their long-term impacts on organizational performance, innovation, and employee well-being (Balasubramanian et al., 2022 ; Basu et al., 2023 ). Finally, further exploration of the AI-Human Decision Matrix through empirical validation is expected. Researchers may explore the boundary conditions under which each decision paradigm is most effective and investigate the factors, such as task complexity, data ambiguity, or cultural variation, that moderate the success of each AI-human collaboration mode. CONCLUSION In this article, we examine the role of AI in human decision-making through the lens of four AI decision-making paradigms: Autonomous Execution, Guided Resolution, Collaborative Exploration, and Augmented Discovery. This paradigm offers a novel perspective for understanding how AI integrates computational capabilities with human creativity across diverse decision-making contexts, particularly in closed and open decision environments. It highlights the dual role of AI in human decision-making: functioning independently to lead complex decision processes and empowering humans by enhancing their performance in information analysis and creative exploration, thereby advancing decision efficiency and fostering innovation. This approach provides a strategic framework for the organizational application of AI while emphasizing the importance of transparency, accountability, and interdisciplinary integration. Future research should further investigate the long-term implications of AI deployment, the dynamics of team collaboration, and the development of regulatory frameworks to enable the responsible and effective use of AI in complex and dynamic environments. By integrating theory and practice, this framework aim f to offer fresh insights for both academic inquiry and industry application, contributing to AI-driven decision-making innovation and promoting sustainable development. Declarations Author Contribution H.L. conducted the literature search, performed data analysis, and drafted the initial manuscript. F.T. provided conceptual guidance, supervised the research process, and critically revised the manuscript. All authors read and approved the final manuscript and agree to be accountable for all aspects of the work. References Abbasi, M. F., Bilal, M., & Rasheed, K. (2022). Role of Human Intuition in AI Aided Managerial Decision Making: A Review. 2022 International Conference on Decision Aid Sciences and Applications (DASA) , 713–718. Afuah, A., & Tucci, C. L. (2003). A model of the Internet as creative destroyer. IEEE Transactions on Engineering Management , 50(4), 395–402. Ahmed, A., Bhatti, S. H., Gölgeci, I., & Arslan, A. (2022). Digital platform capability and organizational agility of emerging market manufacturing SMEs: The mediating role of intellectual capital and the moderating role of environmental dynamism. Technological Forecasting and Social Change , 177, 121513. Akhtar, P., Frynas, J. G., Mellahi, K., & Ullah, S. (2019). Big Data-Savvy Teams’ Skills, Big Data-Driven Actions and Business Performance. British Journal of Management , 30(2), 252–271. Akhtar, P., Khan, Z., Frynas, J. G., Tse, Y. K., & Rao-Nicholson, R. (2019). Essential micro-foundations for contemporary business operations: Top management tangible competencies, relationship-based business networks and environmental sustainability. British Journal of Management , 30(3), 610-629. Allal-Chérif, O., Yela Aránega, A., & Castaño Sánchez, R. (2021). Intelligent recruitment: How to identify, select, and retain talents from around the world using AI. Technological Forecasting and Social Change , 169, 120822. Allen, M. (2017). The SAGE Encyclopedia of Communication Research Methods. SAGE Publications , Incorporated. Anim-Yeboah, S., Boateng, R., Odoom, R., & Kolog, E. A. (2020). Digital Transformation Process and the Capability and Capacity Implications for Small and Medium Enterprises. International Journal of E-Entrepreneurship and Innovation (IJEEI) , 10(2), 26–44. Arbnor, I., & Bjerke, B. (2009). Methodology for Creating Business Knowledge. SAGE Publications , Limited. Arslan, A., Tarba, S. Y., & Larimo, J. (2022). The role of AI in international business: A review and research agenda. International Business Review , 31(1), 101-115. Asokan, D. R., Huq, F. A., Smith, C. M., & Stevenson, M. (2022). Socially responsible operations in the Industry 4.0 era: Post-COVID-19 technology adoption and perspectives on future research. International Journal of Operations & Production Management , 42(13), 185–217. Astaneh Asl, B., & Dossick, C. S. (2022). Immersive VR versus BIM for AEC Team Collaboration in Remote 3D Coordination Processes. Buildings , 12(10), 1548. Bakker, A. B., Emmerik, H. van, & Euwema, M. C. (2006). Crossover of Burnout and Engagement in Work Teams. Work and Occupations , 33(4), 464–489. Balasubramanian, N., Ye, Y., & Xu, M. (2022). Substituting Human Decision-Making with Machine Learning: Implications for Organizational Learning. The Academy of Management Review , 47(3), 448–465. Bankins, S., Ocampo, A. C., Marrone, M., Restubog, S. L. D., & Woo, S. E. (2024). A multilevel review of artificial intelligence in organizations: Implications for organizational behavior research and practice. Journal of Organizational Behavior , 45(2), 159–182. Barney, J. B. (2001). Resource-based theories of competitive advantage: A ten-year retrospective on the resource-based view. Journal of Management , 27(6), 643–650. Basu, S., Ghosh, P., & Biswas, S. (2023). The impact of AI on workplace outcomes: A comprehensive review. Journal of Business Research , 145, 123-135. Bates, C., Crum, E., Brooks, D. J., & Andres, G. (2022). Strategic team engagement for successful documentation of pain management. Journal of Clinical Oncology , 40(28_suppl), 53–53. Bates, M. J. (2012). Decision Support Systems. In Understanding Information Retrieval Systems (pp. 481–492) . Auerbach Publications. Bathla, G., Bhadane, K., Singh, R. K., Kumar, R., Aluvalu, R., Krishnamurthi, R., Kumar, A., Thakur, R. N., Basheer, S., Kumar Reddy, M. P., & M Praveen Kumar Reddy. (2022). Autonomous Vehicles and Intelligent Automation: Applications, Challenges, and Opportunities. Mobile Information Systems , 2022, 1–36. Bell, S. T. (2007). Deep-level composition variables as predictors of team performance: A meta-analysis. Journal of Applied Psychology , 92(3), 595–615. Bhattacharjya, D., & Deleris, L. A. (2012). From reliability block diagrams to fault tree circuits. Decision Analysis , 9(2), 128–137. Blay, T., & Froese, F. J. (2022). The Future of Global Work: Challenges and Recommendations for Global Virtual Teamwork. In H. Merchant (Ed.), The New Frontiers of International Business: Development, Evolving Topics, and Implications for Practice (pp. 237–252) . Borges, A. F. S., Laurindo, F. J. B., Spínola, M. M., Gonçalves, R. F., & Mattos, C. A. (2021). The strategic use of artificial intelligence in the digital era: Systematic literature review and future research directions. International Journal of Information Management , 57 , 102225-. Borissova, D., Dimitrova, Z., & Dimitrov, V. (2020). How to Support Teams to be Remote and Productive: Group Decision-Making for Distance Collaboration Software Tools. Information & Security: An International Journal , 46(1), 36–52. Brielmaier, C., & Friesl, M. (2023). The attention‐based view: Review and conceptual extension towards situated attention. International Journal of Management Reviews , 25(1), 99–129. Brynjolfsson, E., & McAfee, A. (2017). The Business of Artificial Intelligence: What It Can — and Cannot — Do for Your Organization. Harvard Business Review. Brynjolfsson, E., & McAfee, A. (2023). The second machine age: Work, progress, and prosperity in a time of brilliant technologies. W. W. Norton & Company . Bulgakova, I. N., Alexandrova, T. V., & Elokhov, A. M. (2022). DEVELOPMENT OF A METHODOLOGY FOR BUILDING A PROJECT TEAM IN THE HR-MANAGEMENT SYSTEM OF AN ORGANIZATION. Tyumen State University Herald. Social, Economic, and Law Research , 8(1), 269–287. Caputo, F., Keller, B., Moehring, M., Carrubbo, L., & Schmidt, R. (2023). Advancing beyond technicism when managing big data in companies’ decision-making. JOURNAL OF KNOWLEDGE MANAGEMENT . Carmeli, A., Brueller, D., & Dutton, J. E. (2019). Learning Behaviours in the Workplace: The Role of High-quality Interpersonal Relationships and Psychological Safety. Systems Research and Behavioral Science , 26(1), 81-98. Cascio, W. F., & Montealegre, R. (2016). How technology is changing work and organizations. Annual Review of Organizational Psychology and Organizational Behavior , 3(1), 349-375. Chan, D. (1998). Functional relations among constructs in the same content domain at different levels of analysis: A typology of composition models. Journal of Applied Psychology , 83(2), 234–246. Chan, S. H., Song, Q., Sarker, S., & Plumlee, R. D. (2017). Decision support system (DSS) use and decision performance: DSS motivation and its antecedents. Information & Management , 54(7), 934–947. Chatterjee, S., Chaudhuri, R., & Vrontis, D. (2022). AI and digitalization in relationship management: Impact of adopting AI-embedded CRM system. Journal of Business Research , 150, 437–450. Charlwood, A., & Guenole, N. (2022). The role of AI in HRM: A people-centric approach. Human Resource Management Review , 32(2), 100789. Chen, C. (2024). Influence of big data analytical capability on new product performance – the effects of collaboration capability and team collaboration in high-tech firm. Chinese Management Studies , 18(1), 1–23. Chen, G., & Klimoski, R. J. (2003). The Impact of Expectations on Newcomer Performance in Teams as Mediated by Work Characteristics, Social Exchanges, and Empowerment. Academy of Management Journal , 46(5), 591–607. Chernyakhovskaya, L. R., Malakhova, A. I., Nikulina, N. O., & Batalova, V. I. (2021). Information and analytical collective decision-making support using intelligent technologies. Journal of Physics: Conference Series , 1864(1), 012090. Choudhury, P., Foroughi, C., & Larson, B. Z. (2020). Work-from-anywhere: The productivity effects of geographic flexibility. Strategic Management Journal , 41(9), 1641-1664. Choudhury, P., Starr, E., & Agarwal, R. (2020). Machine learning and human capital complementarities: Experimental evidence on bias mitigation. Strategic Management Journal , 41(8), 1381–1411. Chowdhury, S., Dey, P., Joel-Edgar, S., Bhattacharya, S., Rodriguez-Espindola, O., Abadie, A., & Truong, L. (2023). Unlocking the value of artificial intelligence in human resource management through AI capability framework. Human Resource Management Review , 33(1), 100899. Costa, P. L., Passos, A. M., & Bakker, A. B. (2014). Team work engagement: A model of emergence. Journal of Occupational and Organizational Psychology , 87(2), 414–436. Cremona, L., Ravarini, A., & Sutanto, J. (2014). KMS in a Cluster of Firms: The Role of a Digital Platform. In D. Baglieri, C. Metallo, C. Rossignoli, & M. Pezzillo Iacono (Eds.), Information Systems, Management, Organization and Control (pp. 253–262) . Springer International Publishing. Cropanzano, R., & Mitchell, M. S. (2005). Social Exchange Theory: An Interdisciplinary Review. Journal of Management , 31(6), 874–900. Davenport, T. H., & Ronanki, R. (2018). Artificial Intelligence for the Real World. Harvard Business Review , 101(1), 108-116. De Dreu, C. K., & West, M. A. (2001). Minority dissent and team innovation: The importance of participation in decision making. The Journal of Applied Psychology , 86(6), 1191–1201. Deva Hema, D., & Kumar, K. A. (2023). Optimized Deep Neural Network Based Intelligent Decision Support System for Traffic State Prediction. International Journal of Intelligent Transportation Systems Research , 21(1), 26–35. Dvorsky, G. (2013). The 12 cognitive biases that prevent you from being rational. Fan, D., Lau, C. M., & Zhao, X. (2022). The adoption of AI in business: A systematic review and future research agenda. Journal of Business Research , 139, 102-120. Fleming, P. (2019). Robots and Organization Studies: Why Robots Might Not Want to Steal Your Job. Organization Studies , 40(1), 23–38. Fountaine, T., McCarthy, B., & Saleh, T. (2019). Building the AI-powered organization. Harvard Business Review , 97(4), 62-73. Gal, U., Jensen, T. B., & Stein, M.-K. (2020). Breaking the vicious cycle of algorithmic management: A virtue ethics approach to people analytics. Information and Organization, 30(2), 100301. Gilson, L. L., Maynard, M. T., Jones Young, N. C., Vartiainen, M., & Hakonen, M. (2015). Virtual Teams Research: 10 Years, 10 Themes, and 10 Opportunities. Journal of Management , 41 (5), 1313–1337. Gimpel, H., Laubacher, R., Meindl, O., Woehl, M., & Dombetzki, L. (2024). Advancing Content Synthesis in Macro-Task Crowdsourcing Facilitation Leveraging Natural Language Processing. Group Decision And Negotiation , 33(5), 1301–1322. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press . Grigorievich, T. A. (2024). The implementation of lean and digital management techniques using artificial intelligence in industrial settings. Discover Artificial Intelligence , 4(1), 94-. Gursoy, D., Chi, O. H., Lu, L., & Nunkoo, R. (2019). Consumers acceptance of artificially intelligent (AI) device use in service delivery. International Journal of Information Management , 49 , 157–169. Haenlein, M., & Kaplan, A. (2019). A Brief History of AI: On the Past, Present, and Future of AI. California Management Review , 61(4), 5–14. Huang, M.-H., Rust, R., & Maksimovic, V. (2019). The Feeling Economy: Managing in the Next Generation of AI (AI). California Management Review , 61(4), 43–65. Huang, M. H., & Rust, R. T. (2018). "Artificial intelligence in service." Journal of Service Research , 21(2), 155-172 IBM. (2023). Global AI adoption index 2023. IBM Corporation . Retrieved from https://www.ibm.com/reports/global-ai-adoption-index-2023 Jarrahi, M. H. (2023). Algorithmic management and the future of work: A review and research agenda. Journal of Business Research , 145, 123-135. Jussupow, E., Spohrer, K., Heinzl, A., & Gawlitza, J. (2021). Augmenting Medical Diagnosis Decisions? An Investigation into Physicians’ Decision-Making Process with AI. Information Systems Research , 32(3), 713–735. Kaplan, A., & Haenlein, M. (2023). Siri, Siri, in my hand: Who’s the fairest in the land? On the interpretations, illustrations, and implications of artificial intelligence. Business Horizons , 66(1), 15-25. Kauffmann, E., Peral, J., Gil, D., Ferrández, A., Sellers, R., & Mora, H. (2020). A framework for big data analytics in commercial social networks: A case study on sentiment analysis and fake review detection for marketing decision-making. Industrial Marketing Management , 90, 523–537. Kelan, E. K. (2023). Algorithmic inclusion: Shaping the predictive algorithms of artificial intelligence in hiring. Human Resource Management Journal . Kellogg, K. C., Valentine, M. A., & Christin, A. (2020). Algorithms at Work: The New Contested Terrain of Control. Academy of Management Annals , 14(1), 366–410. Kim, J., Lee, S., & Hwang, Y. (2021). A bird’s-eye view of AI research: A comprehensive literature review. Journal of Information Technology , 36(2), 123-135. Köchling, A., & Wehner, M. C. (2023). Better explaining the benefits why AI? Analyzing the impact of explaining the benefits of AI-supported selection on applicant responses. International Journal of Selection and Assessment , 31(1), 45–62. Leavitt, K., Schabram, K., Hariharan, P., & Barnes, C. M. (2021). Ghost in the Machine: On Organizational Theory in the Age of Machine Learning. Academy of Management Review , 46(4), 750–777. Lee, M. C. M., Scheepers, H., Lui, A. K. H., & Ngai, E. W. T. (2023). The implementation of artificial intelligence in organizations: A systematic literature review. Information & Management , 60(5), 103816. Lin, R., Gal, Y., Kraus, S., & Mazliah, Y. (2014). Training with automated agents improves people’s behavior in negotiation and coordination tasks. Decision Support Systems , 60, 1–9. Lindsay, C., Commander, J., Findlay, P., Bennie, M., Dunlop Corcoran, E., & Van Der Meer, R. (2014). ‘Lean’, new technologies and employment in public health services: Employees’ experiences in the National Health Service. The International Journal of Human Resource Management , 25(21), 2941–2956. Lu, J., Yan, Y., Huang, K., Yin, M., & Zhang, F. (2025). Do We Learn From Each Other: Understanding the Human-AI Co-Learning Process Embedded in Human-AI Collaboration. Group Decision And Negotiation , 34 (2), 235–271. McKinsey Global Institute. (2018). Notes from the AI frontier: Insights from hundreds of use cases. Metcalf, L., Askay, D. A., & Rosenberg, L. B. (2019). Keeping Humans in the Loop: Pooling Knowledge through Artificial Swarm Intelligence to Improve Business Decision Making. California Management Review , 61(4), 84–109. Newman, D. T., Fast, N. J., & Harmon, D. J. (2020). When eliminating bias isn’t fair: Algorithmic reductionism and procedural justice in human resource decisions. Organizational Behavior and Human Decision Processes , 160, 149–167. Ng, A. Y. (2016). Machine learning and AI in healthcare. Nature Medicine , 22(1), 23-30. Oracle and Future Workplace. (2019). AI at work: The future of artificial intelligence in the workplace. Oracle Corporation. Pajasmaa, J., Miettinen, K., & Silvennoinen, J. (2025). Group Decision Making in Multiobjective Optimization: A Systematic Literature Review. Group Decision and Negotiation , 34 (2), 329–371. Pan, S. L., & Froese, F. J. (2023). The role of AI in organizational transformation: A review and research agenda. Journal of Business Research , 145, 123-135. Pessach, D., Singer, G., Avrahami, D., Chalutz Ben-Gal, H., Shmueli, E., & Ben-Gal, I. (2020). Employees recruitment: A prescriptive analytics approach via machine learning and mathematical programming. Decision Support Systems , 134, 113290. Pietronudo, M. C., Croidieu, G., & Schiavone, F. (2022). A solution looking for problems? A systematic literature review of the rationalizing influence of artificial intelligence on decision-making in innovation management. Technological Forecasting & Social Change , 182, 121828-. PwC. (2017). Sizing the prize: What’s the real value of AI for your business and how can you capitalise? Raisch, S., & Fomina, K. (2023). Combining Human and Artificial Intelligence: Hybrid Problem-Solving in Organizations. The Academy of Management Review , 1-. Raisch, S., & Krakowski, S. (2021). AI and Management: The Automation–Augmentation Paradox. Academy of Management Review , 46(1), 192–215. Rane, S. B., & Narvel, Y. A. M. (2022). Data-driven decision making with Blockchain-IoT integrated architecture: a project resource management agility perspective of industry 4.0. International Journal of System Assurance Engineering and Management , 13(2), 1005-1023. Russell, S., & Norvig, P. (2022). Artificial intelligence: A modern approach (4th ed.). Pearson. Sahoo, S., & Lo, C. Y. (2022). Smart manufacturing powered by recent technological advancements: A review. Journal of Manufacturing Systems , 64, 236-250. Sajjadiani, S., Sojourner, A. J., Kammeyer-Mueller, J. D., & Mykerezi, E. (2019). Using machine learning to translate applicant work history into predictors of performance and turnover. Journal of Applied Psychology , 104(10), 1207–1225. Serradilla, O., Zugasti, E., Rodriguez, J., & Zurutuza, U. (2022). Deep learning models for predictive maintenance: a survey, comparison, challenges and prospects. Applied Intelligence , 52(10), 10934-10964. Shrestha, Y. R., Ben-Menahem, S. M., & von Krogh, G. (2019). Organizational Decision-Making Structures in the Age of Artificial Intelligence. California Management Review , 61(4), 66–83. Sleep, L. N. (2022). From Making Automated Decision Making Visible to Mapping the Unknowable Human: Counter-Mapping Automated Decision Making in Social Services in Australia. Qualitative Inquiry , 28(7), 848–858. Tambe, P., Cappelli, P., & Yakubovich, V. (2019). Artificial Intelligence in Human Resources Management: Challenges and a Path Forward. California Management Review , 61(4), 15–42. Tay, L., Ng, V., Malik, A., Zhang, J., Chae, J., Ebert, D. S., Ding, Y., Zhao, J., & Kern, M. (2018). Big Data Visualizations in Organizational Science. Organizational Research Methods , 21(3), 660–685. Townsend, D. M., Hunt, R. A., Rady, J., Manocha, P., & Jin, J. hyeong. (2024). Are the Futures Computable? Knightian Uncertainty and Artificial Intelligence. The Academy of Management Review . Tranfield, D., Denyer, D., & Smart, P. (2003). Towards a methodology for developing evidence-informed management knowledge by means of systematic review. British Journal of Management , 14(3), 207-222. Upadhyay, A. K., & Khandelwal, K. (2018). Applying artificial intelligence: implications for recruitment. Strategic HR Review , 17(5), 255-258. Vrontis, D., Christofi, M., Pereira, V., Tarba, S., Makrides, A., & Trichina, E. (2022). Artificial intelligence, robotics, advanced technologies and human resource management: a systematic review. International Journal of Human Resource Management , 33(6), 1237–1266. Vrontis, D., Thrassou, A., & Amirkhanpour, M. (2022). B2B technology marketing: The role of AI in decision-making processes. Industrial Marketing Management , 101, 123-135. Waldkirch, M., Bucher, E., Schou, P. K., & Gruenwald, E. (2021). Controlled by the algorithm, coached by the crowd—How HRM activities take shape on digital work platforms in the gig economy. International Journal of Human Resource Management , 32(12), 2643–2682. Wirtz, J., Patterson, P. G., Kunz, W. H., Gruber, T., Lu, V. N., Paluch, S., & Martins, A. (2018). "Brave new world: service robots in the frontline." Journal of Service Management , 29(5), 907-931. Wyssenbach, T., Zeballos, M., Loosli, S., & Schwaninger, A. (2022). Nonverbal behavior of interviewers influences the competence ratings of observers in recruitment interviews: a study investigating social influence using 360-degree videos with virtual reality and 2D screen displays. Virtual Reality , 26 (2), 669-686. Zhou, Y., Yang, C., Liu, Z., & Gong, L. (2023). Digital technology adoption and innovation performance: A moderated mediation model. Technology Analysis & Strategic Management , 1-16. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 03 Apr, 2026 Read the published version in Group Decision and Negotiation → 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-6885768","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":500516620,"identity":"f8735784-2e7a-463f-b300-27d41fb0277b","order_by":0,"name":"HAN LI","email":"","orcid":"","institution":"University of Newcastle Australia","correspondingAuthor":false,"prefix":"","firstName":"HAN","middleName":"","lastName":"LI","suffix":""},{"id":500516625,"identity":"99c522aa-79fc-477e-b88d-3e2f4df2f6bb","order_by":1,"name":"Feng Tian","email":"data:image/png;base64,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","orcid":"","institution":"University of Newcastle Australia","correspondingAuthor":true,"prefix":"","firstName":"Feng","middleName":"","lastName":"Tian","suffix":""}],"badges":[],"createdAt":"2025-06-13 07:38:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6885768/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6885768/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10726-026-09980-1","type":"published","date":"2026-04-03T15:59:11+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":89281981,"identity":"b328a0c6-e3f7-4b31-9ff9-4e43b136dad8","added_by":"auto","created_at":"2025-08-18 10:39:19","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":442173,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePRISMA Process\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6885768/v1/46ca02fcd8b784c00c2e7ca7.jpeg"},{"id":89281991,"identity":"a056c7e0-4068-4381-a9c4-e6c0cc85afb2","added_by":"auto","created_at":"2025-08-18 10:39:19","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1086585,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKeywords Network\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6885768/v1/fe9f0cbecaada2f76dc3e12e.png"},{"id":89281980,"identity":"90d0d45a-751f-4c5e-a71a-e73ac95e26bc","added_by":"auto","created_at":"2025-08-18 10:39:19","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":237517,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCo-authorship Network\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6885768/v1/9a094b88c6676cdef960562f.jpeg"},{"id":89281985,"identity":"9ead4c5e-0347-4bb6-8466-aab92b2da686","added_by":"auto","created_at":"2025-08-18 10:39:19","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":127497,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHuman-AI Decision-making Matrix\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6885768/v1/eac3e3db6f37ff9325c4011c.png"},{"id":106344416,"identity":"fcf5c238-f6a6-4c23-bd0a-d6635c869b52","added_by":"auto","created_at":"2026-04-07 16:14:20","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2608773,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6885768/v1/cca5aa6f-2d8c-40d5-9484-4764b72c2d6f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Advancing Decision-Making through AI-Human Collaboration: A Systematic Review and Conceptual Framework","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eArtificial intelligence (AI) is recognized as a transformative force reshaping employees, teams, and organizations (Cascio \u0026amp; Montealegre, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). AI decision-making is defined as the process of using artificial intelligence (AI) to support, enhance, or revolutionize the decision-making process, effectively incorporating AI technology into the decision-making process (Russell \u0026amp; Norvig, \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Brynjolfsson \u0026amp; McAfee, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). This new decision-making framework includes leveraging various AI technologies such as machine learning, natural language processing, and data analytics to process and analyze large volumes of data, identify patterns, and generate actionable insights (Goodfellow, Bengio, \u0026amp; Courville, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; McKinsey Global Institute, \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Ng, \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Kim, Lee, \u0026amp; Hwang, \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Lee et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eOrganizations are increasingly adopting AI to enhance decision quality, driven by demands for speed, accuracy, and adaptability in volatile environments (Akhtar et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Vrontis et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). For example, AI applications range from machine learning algorithms predicting market trends (Fountaine et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) and AI-driven analytics optimizing supply chain operations (Choudhury et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), to AI-based tools supporting financial decision-making (Davenport \u0026amp; Ronanki, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). A study conducted by Oracle and Future Workplace (\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) revealed a growing influence of AI, with 50 percent of employees presently applying various forms of AI to assist in decision-making.\u003c/p\u003e\u003cp\u003eDespite these significant advantages, integrating AI into decision-making processes not only promises to boost efficiency and effectiveness but also introduces new complexities and challenges that require a deeper understanding of its mechanisms and outcomes (Choudhury et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The rapid adoption of AI tools has not been as smooth as anticipated (Arslan et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), and the technological inflexibility potentially may lead to organizational disruptions, thereby diminishing the advantages of AI in decision-making and escalating governance costs (Basu et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Given the extensive applications of AI in decision-making, the emerging research field of AI-empowered decision-making has garnered more attention from both scholars and practitioners. To promote academic development in this field, some researchers have attempted to review or analyze relevant literature to guide future research. For example, Borges (2021) identified key sources of value creation, such as decision support, customer engagement, automation, and the development of new products and services, emphasizing AI's potential to transform organizational strategy. Pietronudo (2021) argues that while AI\u0026rsquo;s ability to rationalize decision-making and foster creativity has been acknowledged, its impact on organizational processes remains multifaceted, leading to both expected and unexpected outcomes. At the same time, scholars have proposed hybrid models that combine human intelligence with AI to address complex problems. These models suggest that AI can expand the range of potential solutions, thus complementing human decision-making (Raisch \u0026amp; Fomina, \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). A framework for integrating human and AI decision-making emphasizes key factors such as search space specificity, decision speed, and transparency, while also addressing challenges like AI bias and the need for accountability in algorithmic decision-making (Shrestha et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Additionally, research on AI's impact on the workforce reveals its significant influence on job roles, required skills, and employee attitudes. As AI continues to reshape organizational practices and labor markets, understanding its long-term effects on innovation and workforce dynamics is crucial (Bankins et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eWhile prior studies explore narrow aspects of AI\u0026rsquo;s role, such as automation (Brynjolfsson \u0026amp; McAfee, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), decision support (Shrestha et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), or ethical implications (Raisch \u0026amp; Krakowski, \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), they lack a holistic perspective. Existing reviews either focus on technical capabilities, human-AI interaction models, or organizational impacts in isolation, leaving a fragmented understanding of how these dimensions interact (Raisch \u0026amp; Fomina, \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Bankins et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). This fragmentation obscures the mechanisms through which AI reshapes decision processes and inhibits the development of actionable frameworks for practitioners.\u003c/p\u003e\u003cp\u003eThis study addresses two critical gaps in the existing literature. Theoretically, prior research has suffered from fragmentation due to the absence of integrative frameworks that explain how AI-human collaboration varies across different decision contexts, such as structured versus unstructured problems, and power dynamics, including AI-led versus human-led processes. While several systematic reviews have tackled decision-making in complex contexts such as multiobjective optimization (Pajasmaa et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), few have specifically focused on the evolving dynamics of AI-human collaboration across decision typologies. Practically, organizations face challenges in aligning AI adoption with decision-making needs, often overemphasizing automation at the expense of creativity and ethical considerations. To bridge these gaps, we conducted a systematic review and bibliometric analysis of 1,005 articles, which culminated in the development of a novel AI-Human Decision Matrix. This framework categorizes decision-making into four paradigms, autonomous execution, guided resolution, collaborative exploration, and augmented discovery, based on two dimensions: AI-human dynamics (ranging from dominance to empowerment) and decision typology (spanning open to closed). By mapping these paradigms to various organizational contexts, our study not only provides a strategic tool for optimizing AI adoption but also offers meaningful insights for addressing ethical and operational challenges in the rapidly evolving landscape of decision-making.\u003c/p\u003e\u003cp\u003eTo structure our inquiry, we propose the following research questions: (1) How have scholars conceptualized and classified the roles of AI in organizational decision-making across various contexts? (2) What are the dominant patterns, paradigms, and trends in AI-human collaborative decision-making revealed through bibliometric analysis? (3) How can we develop an integrative framework that maps AI-human dynamics against different decision typologies to inform both academic inquiry and practical application?\u003c/p\u003e\u003cp\u003eBy addressing these questions, this study aims to synthesize fragmented insights into a coherent framework, deepening our understanding of hybrid decision-making systems and informing responsible AI deployment in complex organizational environments.\u003c/p\u003e"},{"header":"METHODOLOGY","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eData Collection\u003c/h2\u003e\u003cp\u003eLiterature data is sourced from the Web of Science (WoS) Core Collection, concentrating on the Social Sciences Citation Index (SSCI) and the Science Citation Index Expanded (SCI-EXPANDED). As highlighted in prior studies (Jiang et al.,2024; Pouris \u0026amp; Pouris, 2011; Mongeon \u0026amp; Paul-Hus, 2016; Borges, 2021; Adriaanse \u0026amp; Rensleigh, 2013), the Web of Science (WoS) database serves as the primary source for data collection. The rationale for selecting the WoS database is multifaceted: First, it encompasses the most significant and influential journals in the field (Pouris \u0026amp; Pouris, 2011). Second, given the interdisciplinary nature of AI Decision-making, which includes information technology and strategic management, WoS is particularly well-suited for our objectives due to its comprehensive coverage across diverse disciplines (Adriaanse \u0026amp; Rensleigh, 2013). Third, WoS is recognized as one of the most prestigious and respected repositories of scientific information in the social sciences (Mongeon \u0026amp; Paul-Hus, 2016), and its utility has been extensively demonstrated in prior review articles related to management decision-making (Borges, 2021).\u003c/p\u003e\u003cp\u003eThe screening process is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. In this study, we focused on journal articles published between 2010 and 2024. The academic literature in the field of AI and Decision-making began in the 1990s, but articles related to AI applications became prominent only after 2010. Therefore, we selected this time frame from 2010 to 2024 to align with the notable growth in AI and machine learning applications in management. The primary search terms were constructed using Boolean operators (AND/OR) to link two groups of keywords. The first group included terms such as \u0026ldquo;decision quality,\u0026rdquo; \u0026ldquo;decision efficiency,\u0026rdquo; \u0026ldquo;decision effectiveness,\u0026rdquo; or simply \u0026ldquo;decision.\u0026rdquo; The second group contained keywords referencing AI-related technologies, such as \u0026ldquo;artificial intelligence,\u0026rdquo; \u0026ldquo;machine learning,\u0026rdquo; or \u0026ldquo;AI.\u0026rdquo; The search was restricted to English-language journal articles, as these are widely regarded as representative of the most influential and reputable research. This process resulted in a total of 28,492 articles.\u003c/p\u003e\u003cp\u003eGiven the multidisciplinary nature of AI research, we refined the results to ensure relevance to management studies. Specifically, we limited the Web of Science categories to include only publications related to management, business, economics, and business finance. This refinement resulted in a total of 1,005 articles.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003eData Analysis\u003c/h2\u003e\u003cp\u003eThe bibliographic records were exported from the Web of Science (WoS) database to VOSviewer 1.6.19 for bibliometric analysis. Compared to other bibliometric tools, VOSviewer offers better visualization capabilities (Van Eck \u0026amp; Waltman, 2010). VOSviewer focuses on the knowledge structure of specific domains, enabling researchers to identify and visualize hidden patterns and trends by extracting information from a large number of bibliographic records (Cobo et al., 2011). Each record includes details such as authors and their affiliations including countries, titles, abstracts, keywords, publication years, source journals, and references. Based on the bibliographic data, analyses of keyword co-occurrence, trend dynamics, and country-wise literature contributions were conducted.\u003c/p\u003e\u003cp\u003e\u003cb\u003eKeyword co-occurrence\u003c/b\u003e\u003c/p\u003e\u003cp\u003eA keyword co-occurrence analysis was conducted to identify relationships among themes within the literature. The co-occurrence network visualization map indicated that these clusters formed a broad and interconnected topic network within the studies (Gupta \u0026amp; Khan, 2024). The software detected links between similar keywords and grouped them into clusters, each represented by a distinct color. Keywords within the same cluster reflect similar or closely related themes (Ali \u0026amp; G\u0026ouml;lgeci, 2019; van Eck \u0026amp; Waltman, 2009). The results of this analysis are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eThe existing themes in AI-related decision-making literature were categorized into four clusters. The red cluster (N\u0026thinsp;=\u0026thinsp;10 terms/keywords) focuses on automated decision-making, comprising research on decision-making. The green cluster (N\u0026thinsp;=\u0026thinsp;11 terms/keywords) is labeled as augmentation and collaboration, comprising research on decision-making. The blue cluster (N\u0026thinsp;=\u0026thinsp;15 terms/keywords) is identified as strategic discovery and innovation. The yellow cluster (N\u0026thinsp;=\u0026thinsp;10 terms/keywords) highlights aspects such as predictive analytics and data analytics.\u003c/p\u003e\u003cp\u003eThe initial red-colored cluster (N\u0026thinsp;=\u0026thinsp;10 terms/keywords) is \u0026ldquo;Autonomous Execution\u0026rdquo; consists of \"Decision-making\", \"Automation\", \"Adoption\" and \"Trust\", which are frequently repeated keywords. In the yellow cluster (N\u0026thinsp;=\u0026thinsp;10 terms/keywords), Guided Resolution, the most frequently repeated keywords are \"Efficiency\", \" Predictive Analytics\" and \" Data Analytics \". The third cluster in green (N\u0026thinsp;=\u0026thinsp;11 terms/keywords) is Collaborative Exploration (AI-dominated Open Decision-Making), composed of the most frequently repeated keywords: \"Augmentation\", \"Creativity\", \"Rational\" and \"Decision support systems\". Lastly, \"Artificial Intelligence (AI)\", \"Innovation\", and \"Complexity\" are the most frequently repeated keywords in the blue cluster (N\u0026thinsp;=\u0026thinsp;15 terms/keywords) Augmented Discovery.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eAuthorship analysis\u003c/h3\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e highlights the leading contributors in articles related to AI decision management, focusing on their publication volume and citation impact. Among the 226 authors analyzed, the ten most highly cited authors are presented, with nine of them receiving over 200 citations. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e demonstrates that Shivam Gupta stands out as the most prolific author, with 10 publications and 457 citations. He is followed by Yogesh Dwivedi, who, despite having published fewer papers (7), achieved the highest citation count of 544. Other notable contributors include Patrick Mikalef (6 publications, 265 citations), Mehrbakhsh Nilashi (5 publications, 281 citations), and Ajay Kumar (5 publications, 274 citations). The remaining authors each have four publications, with Vinit Parida leading this group with 338 citations, followed by Yash Raj Shrestha, Dimitris Bertsimas, Krogh Georg Von, and Arpan Kumar Kar, who demonstrate varying degrees of citation impact.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTop-cited authors for bibliometric reviews\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAuthor\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePapers\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePapers citations\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYogesh Dwivedi\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e544\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eShivam Gupta\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e457\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVinit Parida\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e338\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYash Raj Shrestha\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e322\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDimitris Bertsimas\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e307\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eKrogh Georg Von\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e304\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMehrbakhsh Nilashi\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e281\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAjay Kumar\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e274\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePatrick Mikalef\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e265\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eArpan Kumar Kar\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e193\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eIn addition to citation performance analysis, we employed a co-authorship network mapping to identify the primary groups of contributors to AI decision management literature. The co-authorship network encompasses 21 authors who have each published at least four bibliometric review articles and received at least one citation. This analysis reveals two major clusters, as illustrated in Fig.\u0026nbsp;6, with Gupta Shivam emerging as the central figure in terms of both publication count and citation influence.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003eTo provide an overview of existing research on AI decision-making, we use bibliometric analysis to conduct a comprehensive literature review. Artificial intelligence significantly enhances the speed and effectiveness of decision-making by assisting humans in selecting among various options and generating different alternatives for human choice. Various types of AI decision-making are applied in different social and business contexts. This demonstrates the complementary relationship between humans and AI in the decision-making process (Von Krogh, 2018).\u003c/p\u003e\u003cp\u003eThrough the results of bibliometric analysis, after synthesizing the four clusters presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, we identified and classified an AI decision matrix with the horizontal axis labeled \u0026ldquo;Human-AI Dynamics\u0026rdquo; and the vertical axis labeled \u0026ldquo;Decision-making Paradigm.\u0026rdquo; This matrix allowed us to recognize four distinct AI decision-making types: \"Autonomous Execution,\" which refers to AI-dominated closed decision-making where the system operates independently to reach a conclusion; \"Guided Resolution,\" which refers to AI-empowered closed decision-making where AI supports but does not fully control the decision process; \"Collaborative Exploration,\" which refers to AI-dominated open decision-making where AI leads the exploration and decision-making process in an open-ended context; and \"Augmented Discovery,\" which refers to AI-empowered open decision-making where AI assists human decision-makers in exploring and discovering new possibilities in uncertain environments.\u003c/p\u003e\u003cp\u003eWe defined four types of AI decision-making by using the horizontal axis to represent AI-dominated versus human-dominated decisions, and the vertical axis to represent open versus closed decision-making. The difference among the four AI decision types is based on whether the final decision is made by humans with the support of AI or made by AI with humans\u0026rsquo; supervision, and whether the decision involves a well-defined, structured problem or a diverse, flexible problem with multiple possible answers. By evaluating the effectiveness and applicability of different decision-making models, we can identify scenarios where AI-dominated decisions are more suitable or when human judgment and feedback are necessary. Consequently, we propose four types of AI decision-making: \"Autonomous Execution\" (AI-dominated closed decision-making), \"Guided Resolution\" (AI-empowered closed decision-making), \"Collaborative Exploration\" (AI-dominated open decision-making), and \"Augmented Discovery\" (AI-empowered open decision-making). These types aid in understanding the roles and impacts of different AI decision-making models.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe largest cluster represents the \"Autonomous Execution\" type of AI decision-making, characterized by AI-dominated closed decision-making. The core of this decision-making model lies in delegating certain tasks or operations based on closed decision-making to machines, enabling more comprehensive, rational, and efficient processing. The second largest cluster embodies the \"Guided Resolution\" type of AI decision-making, where AI plays a supportive and enabling role in the decision-making process rather than making decisions autonomously. The third largest cluster represents the \"Collaborative Exploration\" model of AI decision-making. In this model, AI leads the process through alternative decision-making, assuming responsibility for the final decision, while humans provide support and feedback to help AI optimize and adjust decisions. The last cluster represents the \"Augmented Discovery\" type of AI decision-making, focusing on leveraging AI to enhance human capabilities, thereby driving discovery and innovation in open decision-making.\u003c/p\u003e\u003cp\u003eBuilding upon prior literature, we categorize the decision-making process of each type into three distinct stages, problem identification, information collection and processing, and decision implementation (Simon, 1960; March, 1994). These stages serve as a framework to analyze and distinguish the contributions of each study within the clusters, enhancing our ability to relate them to broader decision-making dynamics.\u003c/p\u003e\u003cp\u003eWithin the decision-making process, individuals must navigate complex environments to acquire critical decision-related information, analyze contextual factors and potential outcomes, and determine whether favorable conditions for action exist (Klein, 1998; Gigerenzer \u0026amp; Goldstein, 1996). Significant variation exists among individuals in their understanding and practice of decision-making, driven by factors such as cognitive capabilities (Kahneman, 2011), accumulated experience (Argyris \u0026amp; Sch\u0026ouml;n, 1978), organizational support (Mintzberg et al., 1976), and external environmental uncertainty (Tversky \u0026amp; Kahneman, 1974). The final stage of decision-making is implementation which becomes particularly critical in high-risk and complex scenarios, where effective resource allocation and team collaboration are essential to increase the likelihood of success (Cyert \u0026amp; March, 1963; Bazerman \u0026amp; Moore, 2013). Fully committing to the execution phase fosters the attainment of organizational or individual objectives, enabling broader progress (Gollwitzer, 1999; Heath \u0026amp; Heath, 2010). The different types of AI decision-making exhibit distinct characteristics across the three stages of information collection, information processing, and decision implementation. Below, we describe the features and differences of the four decision-making types. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e summarizes the four clusters and decision types.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eClusters and Decision Types\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDecision Types\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAutonomous Execution (Red)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGuided Resolution (Green)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eCollaborative Exploration (Yellow)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAugmented Discovery (Blue)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eKeywords\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAutomation; Adoption; Decision making; Determinants; Acceptance.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAugmentation; Rational; Collaboration; Uncertainty; Decision support systems.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eEfficiency; Prediction; Simulation; Strategic judgment.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eInnovation; Strategy; Complexity; Discovery; Implementation.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDefinition\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAI-dominated closed decision-making where the system operates independently to reach a conclusion.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAI-empowered closed decision-making where AI supports but does not fully control the decision process\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAI-dominated open decision-making where AI leads the exploration and decision-making process in an open-ended context.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAI-empowered open decision-making where AI assists human decision-makers in exploring and discovering new possibilities in uncertain environments\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInformation Collection\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAI leads the entire data collection process by automating the real-time gathering, processing, and integration of data from various structured sources, ensuring high quality and consistency through techniques like data fusion, monitoring, and cleaning.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAI significantly enhances human decision-making by efficiently processing and analyzing large datasets across various sectors, such as healthcare and finance, thereby extending cognitive capabilities and improving the accuracy and speed of complex analyses.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAI synthesizes data from diverse, unstructured sources to detect trends and shifts, while human input ensures adaptive strategy adjustments, exemplifying the importance of cross-domain data fusion and real-time analytic in gaining competitive advantages.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAI empowers humans by rapidly analyzing complex and unstructured data to uncover latent patterns and emerging trends, significantly enhancing the comprehensiveness, precision, and efficiency of data collection, as demonstrated in social sciences research.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInformation Process\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAI uses advanced techniques to analyze large datasets, identify patterns, make predictions and generate actionable insights.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAI empowers data-driven decision-making by providing analytical insights through its computational power, while humans maintain interpretative control, exemplified in business management and supply chain optimization where AI's analysis supports strategic human judgment.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAI utilizes advanced methodologies such as machine learning, deep learning, and reinforcement learning to uncover latent patterns and create predictive models, enabling actionable forecasts and innovative solutions to open-ended challenges, as demonstrated by Netflix's content strategy optimization.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eThe information processing phase involves intensive collaboration between AI and humans, where AI's predictive modeling is refined by human expertise and creativity, integrating emotional intelligence with data-driven decision-making to foster innovation, as seen in personalized medical treatment planning.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDecision Implementation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAI not only processes information but also autonomously makes and executes decisions with minimal human involvement, ensuring rapid, real-time adjustments and optimizing processes.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAI offers data-driven support, while humans integrate these insights with their expertise and situational judgment, balancing rational and emotional considerations, as seen in healthcare and emergency management contexts.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAI autonomously generates actionable recommendations and leads decision execution, while human oversight ensures iterative refinements, exemplified in new product development where AI designs prototypes and humans validate and adjust outcomes.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eThe decision implementation phase epitomizes human-AI collaboration, where humans integrate AI-generated analyses with intuition, experience, and value-based judgments to execute decisions, exemplified in artistic creation where AI provides inspiration and artists refine outputs to reflect personal style and emotion, showcasing AI-empowered open decision-making.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\n\u003ch3\u003eCluster 1: Autonomous Execution\u003c/h3\u003e\n\u003cp\u003eIn the realm of Autonomous Execution, the role of artificial intelligence is to orchestrate every facet of information collection, processing, and decision implementation. Autonomous Execution can be conceptualized as an AI-dominated closed decision-making process. Within Autonomous Execution, artificial intelligence is tasked not only with acquiring and integrating information from diverse sources but also with conducting in-depth analysis and processing of this information to formulate optimal decisions. At last, AI autonomously implements these decisions based on the analytical outcomes, thereby achieving efficient and precise operations. This comprehensive leadership capability positions artificial intelligence as a pivotal element in Autonomous Execution, ensuring the seamless and effective progression of the entire process.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eInformation Collection Stage\u003c/h2\u003e\u003cp\u003eIn the Autonomous Execution model, the information collection process is entirely led by artificial intelligence (AI). This process typically relies on data from multiple standardized and structured sources, such as sensors, operation logs, enterprise resource planning (ERP) systems, and Internet of Things (IoT) devices(Rane \u0026amp; Narvel, \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). AI automates the real-time collection and processing of this data to ensure high quality and consistency. Specifically, the data collection process often involves multiple technical methods, including data transmission, real-time monitoring, and data cleaning. By utilizing technologies such as Natural Language Processing (NLP), sensor data fusion, and rule-based reasoning engines, AI integrates large volumes of dispersed data to ensure its accuracy and reliability (Balasubramanian et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). For instance, in intelligent manufacturing systems, sensors collect real-time data on equipment parameters such as temperature, vibration, and pressure. AI uses this data to monitor equipment performance and ensure the smooth operation of production processes༈Davenport \u0026amp; Ronanki, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). In this process, AI not only executes data collection but also plays a central role in ensuring the effectiveness of the data, preventing noise or anomalies from compromising the accuracy of decision-making.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eInformation Processing Stage\u003c/h3\u003e\n\u003cp\u003eDuring the information processing stage, it can highlight the core advantages of AI algorithms, including the rapid identification of patterns and trends, as well as the efficient processing of large-scale datasets. AI performs deep processing on the massive data collected to extract useful patterns, trends, or anomalies. AI employs various advanced techniques, including machine learning, deep learning, reinforcement learning, and data mining, to train algorithms and analyze data (Chatterjee \u0026amp; Vrontis, 2022). These techniques help AI identify underlying patterns within the data, make predictions, and recognize patterns. For example, on a production line, AI can use historical data to train models that predict equipment failure times and provide maintenance recommendations (Serradilla et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In the information processing stage, it is characterized by AI's ability to efficiently process data and generate useful information for decision-making through model-based algorithms and data analysis, reflecting the characteristics of rational decision-making (Simon, 1979).\u003c/p\u003e\u003cp\u003eFor example, in autonomous driving, AI processes data from radar and cameras and uses deep neural network models to classify and analyze images, recognize traffic signs, pedestrians, and other vehicles, and predict future driving conditions. These processed results form the basis for AI's decisions, enabling it to respond effectively in complex traffic environments. AI optimizes the decision-making process through algorithms, avoids human intervention, and improves system decisions' efficiency and accuracy (Shrestha et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eDecision Implementation Stage\u003c/h3\u003e\n\u003cp\u003eIn the Autonomous Execution practice, AI not only processes information but also fully takes on the responsibility of executing decisions, with minimal human involvement. In this model, AI automatically implements decisions based on the results of prior information collection and processing, adjusting in real-time according to feedback. For example, in industrial manufacturing, robots automatically adjust machine operations based on AI-generated instructions, optimizing the production process (Grigorievich, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Autonomous vehicles execute braking or steering decisions made by AI to avoid accidents (Bathla et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). At this stage, AI ensures the rapid and efficient execution of decisions through its precise execution capabilities.\u003c/p\u003e\u003cp\u003eIn this process, AI is not only the decision-maker but also the executor of those decisions. The advantage lies in AI's ability to respond instantly, with execution speeds far exceeding human capabilities. Additionally, AI can adjust based on real-time feedback, ensuring precision and consistency throughout the execution process. The decision-execution phase emphasizes minimizing human intervention, reducing human error, and improving the efficiency and accuracy of decision-making and execution (Brynjolfsson \u0026amp; McAfee, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAutonomous Execution further demonstrates that AI systems can address decision-making problems in closed environments, particularly in applications requiring high precision and real-time capabilities. Through information control and execution accuracy, AI maximizes system efficiency. For example, AI systems in fully automated factories can precisely control every detail of the production line, ensuring continuous production and product quality (Sahoo \u0026amp; Lo, \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eCluster 2: Guided Resolution\u003c/h2\u003e\u003cp\u003eIn the Guided Resolution model, artificial intelligence is strategically positioned as a tool and support system for human decision-making, with the primary objective of augmenting human capabilities rather than supplanting them entirely. Guided Resolution can be understood as AI-empowered closed decision-making. Within this framework, artificial intelligence assists by providing functionalities such as data analysis, pattern recognition, and predictive insights, thereby enabling humans to make more effective judgments and actions within closed decision-making environments. The involvement of AI allows humans to better comprehend complex information and contexts, thus enhancing the accuracy and efficiency of decision-making. Through this collaborative approach, artificial intelligence not only elevates human decision-making capabilities but also ensures that humans maintain a leading role throughout the process, fully leveraging their creativity and judgment. This model of human-machine collaboration underscores the value of artificial intelligence as an auxiliary tool, facilitating the maximization of human potential.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eInformation Collection Stage\u003c/h2\u003e\u003cp\u003eDuring the information collection stage, AI aids humans in processing vast amounts of data by filtering, categorizing, and optimizing information, thus providing a clearer foundation for decision-making. For instance, in healthcare, AI supports physicians by analyzing patient health data, such as medical images, genetic test results, or electronic health records, to quickly extract key indicators, thereby accelerating the diagnostic process. This model extends human cognitive and analytical capabilities, demonstrating significant advantages in complex, high-precision environments (Topol, 2019; Daugherty \u0026amp; Wilson, 2018).\u003c/p\u003e\u003cp\u003eIn the financial sector, AI analyzes market trends and risk factors. Natural language processing (NLP) techniques enable AI to rapidly filter financial news for investment-relevant information, optimizing portfolio strategies (Chui et al., 2018). These tools enhance the efficiency of information collection and ensure the reliability of analysis results through automated data cleaning and standardization (Marr, 2021).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eInformation Processing Stage\u003c/h2\u003e\u003cp\u003eDuring information processing, AI leverages its computational power to generate analytical results for human reference. While AI plays a dominant role in analysis, humans retain control over interpreting and utilizing the outcomes. For instance, in business management, AI can analyze customer data to generate recommendations for marketing strategies, such as identifying the most effective advertising times and target audiences (Goodfellow, Bengio, \u0026amp; Courville, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). This process reflects the core principle of Data-Driven Decision-Making, emphasizing the use of data analytics to provide scientific support for decision-making (Brynjolfsson \u0026amp; McAfee, 2014).\u003c/p\u003e\u003cp\u003eIn supply chain management, AI employs optimization algorithms and predictive models to help businesses accurately estimate inventory needs and optimize logistics routes, reducing operational costs. This model combines AI's analytical capabilities with human strategic judgment. For example, under conditions of high market uncertainty, humans may adjust production plans based on AI's forecasts to mitigate potential risks (Agrawal et al., 2018; McKinsey Global Institute, 2020).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eDecision Implementation Stage\u003c/h2\u003e\u003cp\u003eIn the final stage of decision-making and implementation, AI provides auxiliary support, while humans make decisions by integrating AI-generated insights with their own knowledge and situational judgment. For example, in healthcare, physicians may use AI-generated treatment plans as a reference but adapt them during implementation based on patient conditions, personal preferences, and ethical considerations. This approach integrates Rational Decision-Making with Emotional Decision-Making, where AI provides data-driven support, while humans evaluate the appropriateness of decisions using ethics and professional expertise (Lerner et al., 2015; Topol, 2019).\u003c/p\u003e\u003cp\u003eIn emergency management, AI can analyze real-time data to suggest optimal response strategies, such as resource allocation during natural disasters. However, the final action plan requires human deliberation, considering social factors, political decisions, and evolving on-the-ground circumstances (Brynjolfsson et al., 2017; Edmondson \u0026amp; Harvey, 2018).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eCluster 3: Collaborative Exploration\u003c/h2\u003e\u003cp\u003eIn the Collaborative Exploration process, artificial intelligence assumes a leading role in addressing open-ended and highly complex decision-making problems. Collaborative Exploration can be understood as AI-dominated open decision-making. Its defining feature lies in its capacity to conduct exploratory analyses to uncover unknown patterns or generate innovative solutions. Within this model, artificial intelligence aids in identifying potential opportunities and challenges by processing vast amounts of data and performing complex predictive modeling. This approach is particularly well-suited for contexts requiring extensive data processing and intricate predictive modeling, such as product development, new market exploration, or hypothesis generation in scientific research (Kaplan \u0026amp; Haenlein, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Fleming, 2018). In these domains, artificial intelligence not only rapidly analyzes and integrates data from multiple sources but also identifies hidden trends and relationships, offering unprecedented insights and strategies. Through this mechanism, artificial intelligence enhances the innovativeness and effectiveness of decision-making within the Collaborative Exploration model, thereby advancing organizational adaptability and competitive advantage under conditions of uncertainty and complexity.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eInformation Collection Stage\u003c/h2\u003e\u003cp\u003eDuring the information collection stage, AI extracts data from diverse and unstructured sources to construct a comprehensive informational foundation. By integrating data across domains, AI can quickly detect market trends and shifts. For example, natural language processing techniques are employed to analyze social media content, providing insights into consumer behavior(Davenport \u0026amp; Guha, 2020). Meanwhile, human input remains vital during this phase, enabling adaptive adjustments to data strategies in response to evolving demands (Chesbrough, 2003). In drug discovery, for instance, AI can synthesize genomic data, experimental findings, and academic literature to identify potential drug targets and therapeutic pathways (Fleming, 2018). This phase emphasizes the importance of cross-domain data fusion and real-time analytics, which collectively enable organizations to build competitive advantages in complex environments.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003eInformation Processing Stage\u003c/h2\u003e\u003cp\u003eAI employs advanced methodologies, including machine learning, deep learning, and reinforcement learning, to extract deeper insights and construct predictive models (Chatterjee \u0026amp; Vrontis, 2022). The goal is to identify latent patterns and generate actionable forecasts. For instance, in Netflix\u0026rsquo;s content strategy, AI analyzes user viewing preferences, historical consumption data, and cultural trends to forecast the most promising combinations of scripts, directors, and actors, thereby optimizing resource allocation for content production (Westcott Grant, 2018). This exploratory analytical capability allows AI to generate novel insights and provide solutions to open-ended challenges.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003eDecision Implementation Stage\u003c/h2\u003e\u003cp\u003eIn the implementation phase, AI takes on a proactive role by generating actionable recommendations and often leading the execution of final decisions. Nonetheless, human oversight and iterative refinement remain essential to ensure contextual relevance, ethical alignment, and adaptive flexibility. For instance, in new product development, AI can autonomously generate prototype designs aligned with customer preferences or offer conceptual blueprints that serve as a foundation for refinement by R\u0026amp;D teams. Within this decision-making configuration, AI drives the exploratory and generative tasks, while humans focus on validating, interpreting, and tailoring outputs to organizational objectives (Kaplan \u0026amp; Haenlein, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Emerging studies further underscore the significance of mutual learning between human and AI agents in such settings, revealing how co-learning dynamics not only enhance outcome quality but also shape the evolving structure of collaboration within AI-augmented teams (Lu et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003eCluster 4: Augmented Discovery\u003c/h2\u003e\u003cp\u003eAugmented Discovery emphasizes the profound collaboration between artificial intelligence and humans within open decision-making contexts. Augmented Discovery can be understood as AI-empowered open decision-making. In this model, artificial intelligence, through its formidable computational power and data analytical support, assists humans in addressing complex problems while preserving the central role of human creativity, ethical considerations, and value judgment. The involvement of AI enables humans to process vast amounts of information and complex variables more effectively, thereby playing a crucial role in domains such as scientific research, artistic creation, and the resolution of intricate societal challenges (Topol, 2019; Lerner et al., 2015; Deranty \u0026amp; Corbin, 2024). In these areas, artificial intelligence not only provides technical support but also aids in identifying potential patterns and trends, helping humans explore new possibilities and solutions. Through this collaborative approach, the Augmented Discovery model enhances the depth and breadth of decision-making while ensuring human agency and responsibility in the face of uncertainty and complexity. This model of human-machine collaboration fosters innovation and societal progress while maintaining ethical and value-oriented human guidance in the decision-making process.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003eInformation Collection Stage\u003c/h2\u003e\u003cp\u003eDuring the information collection stage, AI aids humans in extracting insights from diverse data sources. The primary objective of this phase is to explore uncharted territories and uncover latent patterns. By analyzing complex and unstructured data, such as texts and images, AI helps identify emerging trends. In the field of social sciences, for instance, AI can analyze vast amounts of social media content to reveal intricate patterns of social behavior or the impacts of public policies (Deranty \u0026amp; Corbin, 2024). This phase exemplifies the empowering role of AI in data collection, wherein its rapid analytical capabilities significantly enhance the comprehensiveness, precision, and efficiency of the process.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003eInformation Processing Stage\u003c/h2\u003e\u003cp\u003eThe information processing stage exemplifies a deeply synergistic interaction between artificial intelligence and human judgment. In this phase, AI employs predictive modeling techniques to produce preliminary insights, which are then enriched and contextualized by human experts drawing on their domain-specific knowledge and creative reasoning. This co-active process blends emotional intelligence, human creativity, and algorithmic precision, thereby enabling data-driven decisions that are both analytically robust and contextually nuanced (Shrestha et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Recent research further highlights the role of natural language processing in supporting complex knowledge synthesis within collaborative environments, enhancing the efficiency and integration of decision-relevant insights in AI-supported settings (Gimpel et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). For instance, in healthcare, AI systems analyze genomic profiles and patient records to propose individualized treatment options, which physicians then refine by considering clinical realities and patient values to arrive at the final therapeutic decision (Topol, 2019).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003eDecision Implementation Stage\u003c/h2\u003e\u003cp\u003eThe decision implementation phase represents the pinnacle of human-AI collaboration. In this stage, humans synthesize AI-generated analyses with their intuition, experience, and value-based judgments to finalize and execute decisions. In the realm of artistic creation, for instance, AI can provide artists with creative inspiration by generating drafts of abstract paintings, which are further refined by artists to reflect their personal style and emotional expression. This process highlights the fusion of creativity and emotion, demonstrating the advantages of AI-empowered open decision-making (Lerner et al., 2015).\u003c/p\u003e\u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThis study utilizes bibliometric analysis to systematically review the impact of AI on decision-making processes. By leveraging bibliometric methods, key themes and trends within the established academic literature are identified and analyzed.. The analysis has identified four critical themes that characterize AI's role in decision-making: Autonomous Execution, Guided Resolution, Collaborative Exploration, and Augmented Discovery. Building on these themes, a novel framework is proposed, categorizing AI's roles as either enabling or leading within structured (closed) and unstructured (open) decision-making environments.. The integration of AI into decision-making has given rise to a variety of models, each characterized by distinct strengths and limitations.\u003c/p\u003e\u003cp\u003eAutonomous Execution demonstrates AI's ability to lead the entire decision-making process, encompassing data collection, analysis, and execution. It is particularly effective in closed decision environments, where AI minimizes human involvement, delivering efficient and accurate decisions while enhancing system automation and intelligence. This approach is widely applicable across industries requiring consistency and speed, such as manufacturing, logistics, and predictive maintenance. In contrast to Autonomous Execution, Guided Resolution focuses on collaboration, leveraging AI's computational and analytical strengths to support human decision-makers while retaining human authority.. This complementary approach is particularly beneficial in domains such as healthcare, finance, supply chain management, and emergency response. By integrating AI's precision with human intuition and ethical reasoning, the Guided Resolution model addresses complex, closed decision scenarios that require a balance of data-driven insights and contextual judgment.\u003c/p\u003e\u003cp\u003eCollaborative Exploration emphasizes AI's dominant role in open-ended decision-making processes, leveraging its ability to integrate vast datasets and generate innovative solutions. Its application is particularly relevant in scenarios requiring exploration and adaptability, such as scientific research, product innovation, and strategic planning. However, AI's limitations in addressing uncertainties and value-based judgments highlight the need for human involvement in later stages to validate and refine AI-generated insights (Brynjolfsson \u0026amp; McAfee, 2014; Raisch \u0026amp; Krakowski, \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAugmented Discovery highlights a co-creative process where AI complements human creativity and expertise, making it ideal for open, dynamic environments. For example, during global health crises, AI can rapidly analyze transmission models and propose mitigation strategies, while human policymakers evaluate these recommendations through ethical and societal lenses (Davenport \u0026amp; Kirby, 2016). Despite its adaptability, challenges such as over-reliance on AI-generated suggestions and concerns about algorithmic transparency and ethics remain critical considerations (Raisch \u0026amp; Krakowski, \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eBuilding on these observations, our study yields important implications for both theory and practice.\u003c/p\u003e\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\u003ch2\u003eTheoretical Implications\u003c/h2\u003e\u003cp\u003eThis study makes several contributions to the theoretical understanding of AI decision-making.\u003c/p\u003e\u003cp\u003eFirst, by proposing the AI-Human Decision Matrix, we categorize AI\u0026rsquo;s roles in decision-making as either enabling or leading, applied across both structured (closed) and unstructured (open) decision contexts. This framework synthesizes previously fragmented literature and clarifies how AI can complement or substitute human involvement at different stages of decision-making (Akhtar et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Raisch \u0026amp; Krakowski, \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSecond, the framework advances classical decision-making theories (e.g., Simon, 1960; March, 1994) by introducing a hybrid lens through which to understand AI\u0026rsquo;s impact on problem identification, information processing, and execution. It extends prior conceptualizations by distinguishing between AI-empowered and AI-dominated roles and mapping these onto specific decision paradigms such as Collaborative Exploration or Augmented Discovery.\u003c/p\u003e\u003cp\u003eThird, our study opens new avenues for theorizing the cognitive division of labor between humans and machines. It reframes AI not merely as a tool for automation or support but also as a potential co-creator of strategic insight in open decision environments (Fleming, 2018; Brynjolfsson \u0026amp; McAfee, 2014).\u003c/p\u003e\u003cp\u003eFinally, the findings underscore critical unresolved theoretical issues that merit further exploration: the need for interdisciplinary integration, longitudinal studies, contextualized analyses, and more nuanced models of human-AI collaboration, particularly as they relate to ethical norms and organizational governance (Arslan et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Raisch \u0026amp; Krakowski, \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cdiv id=\"Sec25\" class=\"Section3\"\u003e\u003ch2\u003ePractical Implications\u003c/h2\u003e\u003cp\u003eFrom a managerial perspective, the framework offers a practical roadmap for aligning AI deployment with organizational decision complexity.\u003c/p\u003e\u003cp\u003eFirst, it enables organizations to assess the fit between AI capabilities and decision typologies. By understanding when to apply AI-led automation versus AI-empowered augmentation, managers can optimize efficiency while maintaining human oversight and creativity. This is particularly important given the trade-offs each model entails, for instance, between speed and transparency, or precision and adaptability.\u003c/p\u003e\u003cp\u003eSecond, the framework informs the design of governance mechanisms that distinguish between decision contexts requiring AI leadership and necessitating human control. Such differentiation is essential for managing risk, ensuring accountability, and fostering trust in AI-augmented systems (Davenport \u0026amp; Ronanki, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Kaplan \u0026amp; Haenlein, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThird, the matrix has implications for capability development. As AI becomes more deeply embedded in strategic and operational decision-making, organizations must equip their workforce with not just data literacy but also collaborative fluency, the ability to interact effectively with AI tools in both closed and open problem spaces. AI integration requires a balanced approach that enhances decision performance without compromising values such as fairness, transparency, and accountability (Brynjolfsson \u0026amp; McAfee, 2014; Fleming, 2018)..\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec26\" class=\"Section3\"\u003e\u003ch2\u003eFuture Research\u003c/h2\u003e\u003cp\u003eLed by the insights generated through our framework and literature analysis, we identify three key directions for future research.\u003c/p\u003e\u003cp\u003eFirst, team-level AI-enabled decision-making remains an underexplored domain. While existing research has predominantly centered on individual or organizational decision contexts, there is a pressing need to investigate how AI integration influences group dynamics, including trust formation, coordination mechanisms, and collective performance (Cascio \u0026amp; Montealegre, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Carmeli et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In this regard, recent empirical work by Lu et al. (\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) highlights the pivotal role of human-AI co-learning in shaping team cognition and behavioral adaptation, offering a promising foundation for future inquiry into hybrid team functioning. Particular emphasis should be placed on understanding how AI affects psychological safety, role allocation, and the development of shared mental models within decision-making teams.\u003c/p\u003e\u003cp\u003eSecond, longitudinal and cross-contextual studies are needed to understand how AI-human collaboration evolves over time and across industries. These studies could assess the stability of different AI-human decision paradigms, such as Autonomous Execution versus Augmented Discovery, and their long-term impacts on organizational performance, innovation, and employee well-being (Balasubramanian et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Basu et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFinally, further exploration of the AI-Human Decision Matrix through empirical validation is expected. Researchers may explore the boundary conditions under which each decision paradigm is most effective and investigate the factors, such as task complexity, data ambiguity, or cultural variation, that moderate the success of each AI-human collaboration mode.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eIn this article, we examine the role of AI in human decision-making through the lens of four AI decision-making paradigms: Autonomous Execution, Guided Resolution, Collaborative Exploration, and Augmented Discovery. This paradigm offers a novel perspective for understanding how AI integrates computational capabilities with human creativity across diverse decision-making contexts, particularly in closed and open decision environments. It highlights the dual role of AI in human decision-making: functioning independently to lead complex decision processes and empowering humans by enhancing their performance in information analysis and creative exploration, thereby advancing decision efficiency and fostering innovation.\u003c/p\u003e\u003cp\u003eThis approach provides a strategic framework for the organizational application of AI while emphasizing the importance of transparency, accountability, and interdisciplinary integration. Future research should further investigate the long-term implications of AI deployment, the dynamics of team collaboration, and the development of regulatory frameworks to enable the responsible and effective use of AI in complex and dynamic environments. By integrating theory and practice, this framework aim f to offer fresh insights for both academic inquiry and industry application, contributing to AI-driven decision-making innovation and promoting sustainable development.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eH.L. conducted the literature search, performed data analysis, and drafted the initial manuscript. F.T. provided conceptual guidance, supervised the research process, and critically revised the manuscript. All authors read and approved the final manuscript and agree to be accountable for all aspects of the work.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbbasi, M. F., Bilal, M., \u0026amp; Rasheed, K. (2022). Role of Human Intuition in AI Aided Managerial Decision Making: A Review. \u003cem\u003e2022 International Conference on Decision Aid Sciences and Applications (DASA)\u003c/em\u003e, 713\u0026ndash;718. \u003c/li\u003e\n\u003cli\u003eAfuah, A., \u0026amp; Tucci, C. L. (2003). A model of the Internet as creative destroyer. \u003cem\u003eIEEE Transactions on Engineering Management\u003c/em\u003e, 50(4), 395\u0026ndash;402. \u003c/li\u003e\n\u003cli\u003eAhmed, A., Bhatti, S. H., G\u0026ouml;lgeci, I., \u0026amp; Arslan, A. (2022). Digital platform capability and organizational agility of emerging market manufacturing SMEs: The mediating role of intellectual capital and the moderating role of environmental dynamism. \u003cem\u003eTechnological Forecasting and Social Change\u003c/em\u003e, 177, 121513. \u003c/li\u003e\n\u003cli\u003eAkhtar, P., Frynas, J. G., Mellahi, K., \u0026amp; Ullah, S. (2019). Big Data-Savvy Teams\u0026rsquo; Skills, Big Data-Driven Actions and Business Performance. \u003cem\u003eBritish Journal of Management\u003c/em\u003e, 30(2), 252\u0026ndash;271. \u003c/li\u003e\n\u003cli\u003eAkhtar, P., Khan, Z., Frynas, J. G., Tse, Y. K., \u0026amp; Rao-Nicholson, R. (2019). Essential micro-foundations for contemporary business operations: Top management tangible competencies, relationship-based business networks and environmental sustainability. \u003cem\u003eBritish Journal of Management\u003c/em\u003e, 30(3), 610-629.\u003c/li\u003e\n\u003cli\u003eAllal-Ch\u0026eacute;rif, O., Yela Ar\u0026aacute;nega, A., \u0026amp; Casta\u0026ntilde;o S\u0026aacute;nchez, R. (2021). Intelligent recruitment: How to identify, select, and retain talents from around the world using AI. \u003cem\u003eTechnological Forecasting and Social Change\u003c/em\u003e, 169, 120822.\u003c/li\u003e\n\u003cli\u003eAllen, M. (2017). The SAGE Encyclopedia of Communication Research Methods. \u003cem\u003eSAGE Publications\u003c/em\u003e, Incorporated. \u003c/li\u003e\n\u003cli\u003eAnim-Yeboah, S., Boateng, R., Odoom, R., \u0026amp; Kolog, E. A. (2020). Digital Transformation Process and the Capability and Capacity Implications for Small and Medium Enterprises. \u003cem\u003eInternational Journal of E-Entrepreneurship and Innovation (IJEEI)\u003c/em\u003e, 10(2), 26\u0026ndash;44. \u003c/li\u003e\n\u003cli\u003eArbnor, I., \u0026amp; Bjerke, B. (2009). Methodology for Creating Business Knowledge. \u003cem\u003eSAGE Publications\u003c/em\u003e, Limited. \u003c/li\u003e\n\u003cli\u003eArslan, A., Tarba, S. Y., \u0026amp; Larimo, J. (2022). The role of AI in international business: A review and research agenda. \u003cem\u003eInternational Business Review\u003c/em\u003e, 31(1), 101-115.\u003c/li\u003e\n\u003cli\u003eAsokan, D. R., Huq, F. A., Smith, C. M., \u0026amp; Stevenson, M. (2022). Socially responsible operations in the Industry 4.0 era: Post-COVID-19 technology adoption and perspectives on future research. \u003cem\u003eInternational Journal of Operations \u0026amp; Production Management\u003c/em\u003e, 42(13), 185\u0026ndash;217. \u003c/li\u003e\n\u003cli\u003eAstaneh Asl, B., \u0026amp; Dossick, C. S. (2022). Immersive VR versus BIM for AEC Team Collaboration in Remote 3D Coordination Processes. \u003cem\u003eBuildings\u003c/em\u003e, 12(10), 1548. \u003c/li\u003e\n\u003cli\u003eBakker, A. B., Emmerik, H. van, \u0026amp; Euwema, M. C. (2006). Crossover of Burnout and Engagement in Work Teams. \u003cem\u003eWork and Occupations\u003c/em\u003e, 33(4), 464\u0026ndash;489. \u003c/li\u003e\n\u003cli\u003eBalasubramanian, N., Ye, Y., \u0026amp; Xu, M. (2022). Substituting Human Decision-Making with Machine Learning: Implications for Organizational Learning. \u003cem\u003eThe Academy of Management Review\u003c/em\u003e, 47(3), 448\u0026ndash;465.\u003c/li\u003e\n\u003cli\u003eBankins, S., Ocampo, A. C., Marrone, M., Restubog, S. L. D., \u0026amp; Woo, S. E. (2024). A multilevel review of artificial intelligence in organizations: Implications for organizational behavior research and practice. \u003cem\u003eJournal of Organizational Behavior\u003c/em\u003e, 45(2), 159\u0026ndash;182.\u003c/li\u003e\n\u003cli\u003eBarney, J. B. (2001). Resource-based theories of competitive advantage: A ten-year retrospective on the resource-based view. \u003cem\u003eJournal of Management\u003c/em\u003e, 27(6), 643\u0026ndash;650. \u003c/li\u003e\n\u003cli\u003eBasu, S., Ghosh, P., \u0026amp; Biswas, S. (2023). The impact of AI on workplace outcomes: A comprehensive review. \u003cem\u003eJournal of Business Research\u003c/em\u003e, 145, 123-135.\u003c/li\u003e\n\u003cli\u003eBates, C., Crum, E., Brooks, D. J., \u0026amp; Andres, G. (2022). Strategic team engagement for successful documentation of pain management. \u003cem\u003eJournal of Clinical Oncology\u003c/em\u003e, 40(28_suppl), 53\u0026ndash;53. \u003c/li\u003e\n\u003cli\u003eBates, M. J. (2012). Decision Support Systems.\u003cem\u003e In Understanding Information Retrieval Systems (pp. 481\u0026ndash;492)\u003c/em\u003e. Auerbach Publications. \u003c/li\u003e\n\u003cli\u003eBathla, G., Bhadane, K., Singh, R. K., Kumar, R., Aluvalu, R., Krishnamurthi, R., Kumar, A., Thakur, R. N., Basheer, S., Kumar Reddy, M. P., \u0026amp; M Praveen Kumar Reddy. (2022). Autonomous Vehicles and Intelligent Automation: Applications, Challenges, and Opportunities. \u003cem\u003eMobile Information Systems\u003c/em\u003e, 2022, 1\u0026ndash;36. \u003c/li\u003e\n\u003cli\u003eBell, S. T. (2007). Deep-level composition variables as predictors of team performance: A meta-analysis. \u003cem\u003eJournal of Applied Psychology\u003c/em\u003e, 92(3), 595\u0026ndash;615. \u003c/li\u003e\n\u003cli\u003eBhattacharjya, D., \u0026amp; Deleris, L. A. (2012). From reliability block diagrams to fault tree circuits. \u003cem\u003eDecision Analysis\u003c/em\u003e, 9(2), 128\u0026ndash;137.\u003c/li\u003e\n\u003cli\u003eBlay, T., \u0026amp; Froese, F. J. (2022). The Future of Global Work: Challenges and Recommendations for Global Virtual Teamwork.\u003cem\u003e In H. Merchant (Ed.), The New Frontiers of International Business: Development, Evolving Topics, and Implications for Practice (pp. 237\u0026ndash;252)\u003c/em\u003e. \u003c/li\u003e\n\u003cli\u003eBorges, A. F. S., Laurindo, F. J. B., Sp\u0026iacute;nola, M. M., Gon\u0026ccedil;alves, R. F., \u0026amp; Mattos, C. A. (2021). The strategic use of artificial intelligence in the digital era: Systematic literature review and future research directions. \u003cem\u003eInternational Journal of Information Management\u003c/em\u003e, \u003cem\u003e57\u003c/em\u003e, 102225-.\u003c/li\u003e\n\u003cli\u003eBorissova, D., Dimitrova, Z., \u0026amp; Dimitrov, V. (2020). How to Support Teams to be Remote and Productive: Group Decision-Making for Distance Collaboration Software Tools. \u003cem\u003eInformation \u0026amp; Security: An International Journal\u003c/em\u003e, 46(1), 36\u0026ndash;52. \u003c/li\u003e\n\u003cli\u003eBrielmaier, C., \u0026amp; Friesl, M. (2023). The attention‐based view: Review and conceptual extension towards situated attention. \u003cem\u003eInternational Journal of Management Reviews\u003c/em\u003e, 25(1), 99\u0026ndash;129. \u003c/li\u003e\n\u003cli\u003eBrynjolfsson, E., \u0026amp; McAfee, A. (2017). The Business of Artificial Intelligence: What It Can \u0026mdash; and Cannot \u0026mdash; Do for Your Organization. \u003cem\u003eHarvard Business Review.\u003c/em\u003e\u003c/li\u003e\n\u003cli\u003eBrynjolfsson, E., \u0026amp; McAfee, A. (2023). The second machine age: Work, progress, and prosperity in a time of brilliant technologies. \u003cem\u003eW. W. Norton \u0026amp; Company\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eBulgakova, I. N., Alexandrova, T. V., \u0026amp; Elokhov, A. M. (2022). DEVELOPMENT OF A METHODOLOGY FOR BUILDING A PROJECT TEAM IN THE HR-MANAGEMENT SYSTEM OF AN ORGANIZATION. Tyumen State University \u003cem\u003eHerald. Social, Economic, and Law Research\u003c/em\u003e, 8(1), 269\u0026ndash;287. \u003c/li\u003e\n\u003cli\u003eCaputo, F., Keller, B., Moehring, M., Carrubbo, L., \u0026amp; Schmidt, R. (2023). Advancing beyond technicism when managing big data in companies\u0026rsquo; decision-making.\u003cem\u003e JOURNAL OF KNOWLEDGE MANAGEMENT\u003c/em\u003e. \u003c/li\u003e\n\u003cli\u003eCarmeli, A., Brueller, D., \u0026amp; Dutton, J. E. (2019). Learning Behaviours in the Workplace: The Role of High-quality Interpersonal Relationships and Psychological Safety.\u003cem\u003e Systems Research and Behavioral Science\u003c/em\u003e, 26(1), 81-98.\u003c/li\u003e\n\u003cli\u003eCascio, W. F., \u0026amp; Montealegre, R. (2016). How technology is changing work and organizations. \u003cem\u003eAnnual Review of Organizational Psychology and Organizational Behavior\u003c/em\u003e, 3(1), 349-375.\u003c/li\u003e\n\u003cli\u003eChan, D. (1998). Functional relations among constructs in the same content domain at different levels of analysis: A typology of composition models.\u003cem\u003e Journal of Applied Psychology\u003c/em\u003e, 83(2), 234\u0026ndash;246. \u003c/li\u003e\n\u003cli\u003eChan, S. H., Song, Q., Sarker, S., \u0026amp; Plumlee, R. D. (2017). Decision support system (DSS) use and decision performance: DSS motivation and its antecedents. \u003cem\u003eInformation \u0026amp; Management\u003c/em\u003e, 54(7), 934\u0026ndash;947. \u003c/li\u003e\n\u003cli\u003eChatterjee, S., Chaudhuri, R., \u0026amp; Vrontis, D. (2022). AI and digitalization in relationship management: Impact of adopting AI-embedded CRM system. \u003cem\u003eJournal of Business Research\u003c/em\u003e, 150, 437\u0026ndash;450. \u003c/li\u003e\n\u003cli\u003eCharlwood, A., \u0026amp; Guenole, N. (2022). The role of AI in HRM: A people-centric approach. \u003cem\u003eHuman Resource Management Review\u003c/em\u003e, 32(2), 100789.\u003c/li\u003e\n\u003cli\u003eChen, C. (2024). Influence of big data analytical capability on new product performance \u0026ndash; the effects of collaboration capability and team collaboration in high-tech firm. \u003cem\u003eChinese Management Studies\u003c/em\u003e, 18(1), 1\u0026ndash;23. \u003c/li\u003e\n\u003cli\u003eChen, G., \u0026amp; Klimoski, R. J. (2003). The Impact of Expectations on Newcomer Performance in Teams as Mediated by Work Characteristics, Social Exchanges, and Empowerment. \u003cem\u003eAcademy of Management Journal\u003c/em\u003e, 46(5), 591\u0026ndash;607. \u003c/li\u003e\n\u003cli\u003eChernyakhovskaya, L. R., Malakhova, A. I., Nikulina, N. O., \u0026amp; Batalova, V. I. (2021). Information and analytical collective decision-making support using intelligent technologies.\u003cem\u003e Journal of Physics: Conference Series\u003c/em\u003e, 1864(1), 012090. \u003c/li\u003e\n\u003cli\u003eChoudhury, P., Foroughi, C., \u0026amp; Larson, B. Z. (2020). Work-from-anywhere: The productivity effects of geographic flexibility. \u003cem\u003eStrategic Management Journal\u003c/em\u003e, 41(9), 1641-1664.\u003c/li\u003e\n\u003cli\u003eChoudhury, P., Starr, E., \u0026amp; Agarwal, R. (2020). Machine learning and human capital complementarities: Experimental evidence on bias mitigation. \u003cem\u003eStrategic Management Journal\u003c/em\u003e, 41(8), 1381\u0026ndash;1411.\u003c/li\u003e\n\u003cli\u003eChowdhury, S., Dey, P., Joel-Edgar, S., Bhattacharya, S., Rodriguez-Espindola, O., Abadie, A., \u0026amp; Truong, L. (2023). Unlocking the value of artificial intelligence in human resource management through AI capability framework. \u003cem\u003eHuman Resource Management Review\u003c/em\u003e, 33(1), 100899.\u003c/li\u003e\n\u003cli\u003eCosta, P. L., Passos, A. M., \u0026amp; Bakker, A. B. (2014). Team work engagement: A model of emergence.\u003cem\u003e Journal of Occupational and Organizational Psychology\u003c/em\u003e, 87(2), 414\u0026ndash;436. \u003c/li\u003e\n\u003cli\u003eCremona, L., Ravarini, A., \u0026amp; Sutanto, J. (2014). KMS in a Cluster of Firms: The Role of a Digital Platform. In D. Baglieri, C. Metallo, C. Rossignoli, \u0026amp; M. Pezzillo Iacono (Eds.), \u003cem\u003eInformation Systems, Management, Organization and Control (pp. 253\u0026ndash;262)\u003c/em\u003e. Springer International Publishing. \u003c/li\u003e\n\u003cli\u003eCropanzano, R., \u0026amp; Mitchell, M. S. (2005). Social Exchange Theory: An Interdisciplinary Review. \u003cem\u003eJournal of Management\u003c/em\u003e, 31(6), 874\u0026ndash;900. \u003c/li\u003e\n\u003cli\u003eDavenport, T. H., \u0026amp; Ronanki, R. (2018). Artificial Intelligence for the Real World. \u003cem\u003eHarvard Business Review\u003c/em\u003e, 101(1), 108-116.\u003c/li\u003e\n\u003cli\u003eDe Dreu, C. K., \u0026amp; West, M. A. (2001). Minority dissent and team innovation: The importance of participation in decision making. \u003cem\u003eThe Journal of Applied Psychology\u003c/em\u003e, 86(6), 1191\u0026ndash;1201. \u003c/li\u003e\n\u003cli\u003eDeva Hema, D., \u0026amp; Kumar, K. A. (2023). Optimized Deep Neural Network Based Intelligent Decision Support System for Traffic State Prediction. \u003cem\u003eInternational Journal of Intelligent Transportation Systems Research\u003c/em\u003e, 21(1), 26\u0026ndash;35. \u003c/li\u003e\n\u003cli\u003eDvorsky, G. (2013). The 12 cognitive biases that prevent you from being rational.\u003c/li\u003e\n\u003cli\u003eFan, D., Lau, C. M., \u0026amp; Zhao, X. (2022). The adoption of AI in business: A systematic review and future research agenda. \u003cem\u003eJournal of Business Research\u003c/em\u003e, 139, 102-120.\u003c/li\u003e\n\u003cli\u003eFleming, P. (2019). Robots and Organization Studies: Why Robots Might Not Want to Steal Your Job. \u003cem\u003eOrganization Studies\u003c/em\u003e, 40(1), 23\u0026ndash;38.\u003c/li\u003e\n\u003cli\u003eFountaine, T., McCarthy, B., \u0026amp; Saleh, T. (2019). Building the AI-powered organization. \u003cem\u003eHarvard Business Review\u003c/em\u003e, 97(4), 62-73.\u003c/li\u003e\n\u003cli\u003eGal, U., Jensen, T. B., \u0026amp; Stein, M.-K. (2020). Breaking the vicious cycle of algorithmic management: A virtue ethics approach to people analytics.\u003cem\u003e Information and Organization,\u003c/em\u003e 30(2), 100301.\u003c/li\u003e\n\u003cli\u003eGilson, L. L., Maynard, M. T., Jones Young, N. C., Vartiainen, M., \u0026amp; Hakonen, M. (2015). Virtual Teams Research: 10 Years, 10 Themes, and 10 Opportunities. \u003cem\u003eJournal of Management\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(5), 1313\u0026ndash;1337.\u003c/li\u003e\n\u003cli\u003eGimpel, H., Laubacher, R., Meindl, O., Woehl, M., \u0026amp; Dombetzki, L. (2024). Advancing Content Synthesis in Macro-Task Crowdsourcing Facilitation Leveraging Natural Language Processing. \u003cem\u003eGroup Decision And Negotiation\u003c/em\u003e, 33(5), 1301\u0026ndash;1322.\u003c/li\u003e\n\u003cli\u003eGoodfellow, I., Bengio, Y., \u0026amp; Courville, A. (2016). Deep Learning. \u003cem\u003eMIT Press\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eGrigorievich, T. A. (2024). The implementation of lean and digital management techniques using artificial intelligence in industrial settings. \u003cem\u003eDiscover Artificial Intelligence\u003c/em\u003e, 4(1), 94-. \u003c/li\u003e\n\u003cli\u003eGursoy, D., Chi, O. H., Lu, L., \u0026amp; Nunkoo, R. (2019). Consumers acceptance of artificially intelligent (AI) device use in service delivery. \u003cem\u003eInternational Journal of Information Management\u003c/em\u003e, \u003cem\u003e49\u003c/em\u003e, 157\u0026ndash;169. \u003c/li\u003e\n\u003cli\u003eHaenlein, M., \u0026amp; Kaplan, A. (2019). A Brief History of AI: On the Past, Present, and Future of AI. \u003cem\u003eCalifornia Management Review\u003c/em\u003e, 61(4), 5\u0026ndash;14.\u003c/li\u003e\n\u003cli\u003eHuang, M.-H., Rust, R., \u0026amp; Maksimovic, V. (2019). The Feeling Economy: Managing in the Next Generation of AI (AI). \u003cem\u003eCalifornia Management Review\u003c/em\u003e, 61(4), 43\u0026ndash;65.\u003c/li\u003e\n\u003cli\u003eHuang, M. H., \u0026amp; Rust, R. T. (2018). \u0026quot;Artificial intelligence in service.\u0026quot; \u003cem\u003eJournal of Service Research\u003c/em\u003e, 21(2), 155-172\u003c/li\u003e\n\u003cli\u003eIBM. (2023). Global AI adoption index 2023.\u003cem\u003e IBM Corporation\u003c/em\u003e. Retrieved from https://www.ibm.com/reports/global-ai-adoption-index-2023\u003c/li\u003e\n\u003cli\u003eJarrahi, M. H. (2023). Algorithmic management and the future of work: A review and research agenda. \u003cem\u003eJournal of Business Research\u003c/em\u003e, 145, 123-135.\u003c/li\u003e\n\u003cli\u003eJussupow, E., Spohrer, K., Heinzl, A., \u0026amp; Gawlitza, J. (2021). Augmenting Medical Diagnosis Decisions? An Investigation into Physicians\u0026rsquo; Decision-Making Process with AI. \u003cem\u003eInformation Systems Research\u003c/em\u003e, 32(3), 713\u0026ndash;735.\u003c/li\u003e\n\u003cli\u003eKaplan, A., \u0026amp; Haenlein, M. (2023). Siri, Siri, in my hand: Who\u0026rsquo;s the fairest in the land? On the interpretations, illustrations, and implications of artificial intelligence. \u003cem\u003eBusiness Horizons\u003c/em\u003e, 66(1), 15-25.\u003c/li\u003e\n\u003cli\u003eKauffmann, E., Peral, J., Gil, D., Ferr\u0026aacute;ndez, A., Sellers, R., \u0026amp; Mora, H. (2020). A framework for big data analytics in commercial social networks: A case study on sentiment analysis and fake review detection for marketing decision-making. \u003cem\u003eIndustrial Marketing Management\u003c/em\u003e, 90, 523\u0026ndash;537.\u003c/li\u003e\n\u003cli\u003eKelan, E. K. (2023). Algorithmic inclusion: Shaping the predictive algorithms of artificial intelligence in hiring. \u003cem\u003eHuman Resource Management Journal\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eKellogg, K. C., Valentine, M. A., \u0026amp; Christin, A. (2020). Algorithms at Work: The New Contested Terrain of Control. \u003cem\u003eAcademy of Management Annals\u003c/em\u003e, 14(1), 366\u0026ndash;410.\u003c/li\u003e\n\u003cli\u003eKim, J., Lee, S., \u0026amp; Hwang, Y. (2021). A bird\u0026rsquo;s-eye view of AI research: A comprehensive literature review.\u003cem\u003e Journal of Information Technology\u003c/em\u003e, 36(2), 123-135.\u003c/li\u003e\n\u003cli\u003eK\u0026ouml;chling, A., \u0026amp; Wehner, M. C. (2023). Better explaining the benefits why AI? Analyzing the impact of explaining the benefits of AI-supported selection on applicant responses. \u003cem\u003eInternational Journal of Selection and Assessment\u003c/em\u003e, 31(1), 45\u0026ndash;62.\u003c/li\u003e\n\u003cli\u003eLeavitt, K., Schabram, K., Hariharan, P., \u0026amp; Barnes, C. M. (2021). Ghost in the Machine: On Organizational Theory in the Age of Machine Learning. \u003cem\u003eAcademy of Management Review\u003c/em\u003e, 46(4), 750\u0026ndash;777.\u003c/li\u003e\n\u003cli\u003eLee, M. C. M., Scheepers, H., Lui, A. K. H., \u0026amp; Ngai, E. W. T. (2023). The implementation of artificial intelligence in organizations: A systematic literature review. \u003cem\u003eInformation \u0026amp; Management\u003c/em\u003e, 60(5), 103816.\u003c/li\u003e\n\u003cli\u003eLin, R., Gal, Y., Kraus, S., \u0026amp; Mazliah, Y. (2014). Training with automated agents improves people\u0026rsquo;s behavior in negotiation and coordination tasks.\u003cem\u003e Decision Support Systems\u003c/em\u003e, 60, 1\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eLindsay, C., Commander, J., Findlay, P., Bennie, M., Dunlop Corcoran, E., \u0026amp; Van Der Meer, R. (2014). \u0026lsquo;Lean\u0026rsquo;, new technologies and employment in public health services: Employees\u0026rsquo; experiences in the National Health Service. \u003cem\u003eThe International Journal of Human Resource Management\u003c/em\u003e, 25(21), 2941\u0026ndash;2956.\u003c/li\u003e\n\u003cli\u003eLu, J., Yan, Y., Huang, K., Yin, M., \u0026amp; Zhang, F. (2025). Do We Learn From Each Other: Understanding the Human-AI Co-Learning Process Embedded in Human-AI Collaboration. \u003cem\u003eGroup Decision And Negotiation\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(2), 235\u0026ndash;271.\u003c/li\u003e\n\u003cli\u003eMcKinsey Global Institute. (2018). Notes from the AI frontier: Insights from hundreds of use cases.\u003c/li\u003e\n\u003cli\u003eMetcalf, L., Askay, D. A., \u0026amp; Rosenberg, L. B. (2019). Keeping Humans in the Loop: Pooling Knowledge through Artificial Swarm Intelligence to Improve Business Decision Making. \u003cem\u003eCalifornia Management Review\u003c/em\u003e, 61(4), 84\u0026ndash;109.\u003c/li\u003e\n\u003cli\u003eNewman, D. T., Fast, N. J., \u0026amp; Harmon, D. J. (2020). When eliminating bias isn\u0026rsquo;t fair: Algorithmic reductionism and procedural justice in human resource decisions. \u003cem\u003eOrganizational Behavior and Human Decision Processes\u003c/em\u003e, 160, 149\u0026ndash;167.\u003c/li\u003e\n\u003cli\u003eNg, A. Y. (2016). Machine learning and AI in healthcare. \u003cem\u003eNature Medicine\u003c/em\u003e, 22(1), 23-30.\u003c/li\u003e\n\u003cli\u003eOracle and Future Workplace. (2019). AI at work: The future of artificial intelligence in the workplace. Oracle Corporation. \u003c/li\u003e\n\u003cli\u003ePajasmaa, J., Miettinen, K., \u0026amp; Silvennoinen, J. (2025). Group Decision Making in Multiobjective Optimization: A Systematic Literature Review. \u003cem\u003eGroup Decision and Negotiation\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(2), 329\u0026ndash;371. \u003c/li\u003e\n\u003cli\u003ePan, S. L., \u0026amp; Froese, F. J. (2023). The role of AI in organizational transformation: A review and research agenda. \u003cem\u003eJournal of Business Research\u003c/em\u003e, 145, 123-135.\u003c/li\u003e\n\u003cli\u003ePessach, D., Singer, G., Avrahami, D., Chalutz Ben-Gal, H., Shmueli, E., \u0026amp; Ben-Gal, I. (2020). Employees recruitment: A prescriptive analytics approach via machine learning and mathematical programming. \u003cem\u003eDecision Support Systems\u003c/em\u003e, 134, 113290.\u003c/li\u003e\n\u003cli\u003ePietronudo, M. C., Croidieu, G., \u0026amp; Schiavone, F. (2022). A solution looking for problems? A systematic literature review of the rationalizing influence of artificial intelligence on decision-making in innovation management. \u003cem\u003eTechnological Forecasting \u0026amp; Social Change\u003c/em\u003e, 182, 121828-.\u003c/li\u003e\n\u003cli\u003ePwC. (2017). Sizing the prize: What\u0026rsquo;s the real value of AI for your business and how can you capitalise?\u003c/li\u003e\n\u003cli\u003eRaisch, S., \u0026amp; Fomina, K. (2023). Combining Human and Artificial Intelligence: Hybrid Problem-Solving in Organizations.\u003cem\u003e The Academy of Management Review\u003c/em\u003e, 1-.\u003c/li\u003e\n\u003cli\u003eRaisch, S., \u0026amp; Krakowski, S. (2021). AI and Management: The Automation\u0026ndash;Augmentation Paradox.\u003cem\u003e Academy of Management Review\u003c/em\u003e, 46(1), 192\u0026ndash;215.\u003c/li\u003e\n\u003cli\u003eRane, S. B., \u0026amp; Narvel, Y. A. M. (2022). Data-driven decision making with Blockchain-IoT integrated architecture: a project resource management agility perspective of industry 4.0. \u003cem\u003eInternational Journal of System Assurance Engineering and Management\u003c/em\u003e, 13(2), 1005-1023.\u003c/li\u003e\n\u003cli\u003eRussell, S., \u0026amp; Norvig, P. (2022). Artificial intelligence: A modern approach (4th ed.). Pearson.\u003c/li\u003e\n\u003cli\u003eSahoo, S., \u0026amp; Lo, C. Y. (2022). Smart manufacturing powered by recent technological advancements: A review. \u003cem\u003eJournal of Manufacturing Systems\u003c/em\u003e, 64, 236-250.\u003c/li\u003e\n\u003cli\u003eSajjadiani, S., Sojourner, A. J., Kammeyer-Mueller, J. D., \u0026amp; Mykerezi, E. (2019). Using machine learning to translate applicant work history into predictors of performance and turnover.\u003cem\u003e Journal of Applied Psychology\u003c/em\u003e, 104(10), 1207\u0026ndash;1225.\u003c/li\u003e\n\u003cli\u003eSerradilla, O., Zugasti, E., Rodriguez, J., \u0026amp; Zurutuza, U. (2022). Deep learning models for predictive maintenance: a survey, comparison, challenges and prospects. \u003cem\u003eApplied Intelligence\u003c/em\u003e, 52(10), 10934-10964.\u003c/li\u003e\n\u003cli\u003eShrestha, Y. R., Ben-Menahem, S. M., \u0026amp; von Krogh, G. (2019). Organizational Decision-Making Structures in the Age of Artificial Intelligence. \u003cem\u003eCalifornia Management Review\u003c/em\u003e, 61(4), 66\u0026ndash;83.\u003c/li\u003e\n\u003cli\u003eSleep, L. N. (2022). From Making Automated Decision Making Visible to Mapping the Unknowable Human: Counter-Mapping Automated Decision Making in Social Services in Australia. \u003cem\u003eQualitative Inquiry\u003c/em\u003e, 28(7), 848\u0026ndash;858.\u003c/li\u003e\n\u003cli\u003eTambe, P., Cappelli, P., \u0026amp; Yakubovich, V. (2019). Artificial Intelligence in Human Resources Management: Challenges and a Path Forward. \u003cem\u003eCalifornia Management Review\u003c/em\u003e, 61(4), 15\u0026ndash;42.\u003c/li\u003e\n\u003cli\u003eTay, L., Ng, V., Malik, A., Zhang, J., Chae, J., Ebert, D. S., Ding, Y., Zhao, J., \u0026amp; Kern, M. (2018). Big Data Visualizations in Organizational Science. \u003cem\u003eOrganizational Research Methods\u003c/em\u003e, 21(3), 660\u0026ndash;685.\u003c/li\u003e\n\u003cli\u003eTownsend, D. M., Hunt, R. A., Rady, J., Manocha, P., \u0026amp; Jin, J. hyeong. (2024). Are the Futures Computable? Knightian Uncertainty and Artificial Intelligence.\u003cem\u003e The Academy of Management Review\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eTranfield, D., Denyer, D., \u0026amp; Smart, P. (2003). Towards a methodology for developing evidence-informed management knowledge by means of systematic review. \u003cem\u003eBritish Journal of Management\u003c/em\u003e, 14(3), 207-222.\u003c/li\u003e\n\u003cli\u003eUpadhyay, A. K., \u0026amp; Khandelwal, K. (2018). Applying artificial intelligence: implications for recruitment.\u003cem\u003e Strategic HR Review\u003c/em\u003e, 17(5), 255-258.\u003c/li\u003e\n\u003cli\u003eVrontis, D., Christofi, M., Pereira, V., Tarba, S., Makrides, A., \u0026amp; Trichina, E. (2022). Artificial intelligence, robotics, advanced technologies and human resource management: a systematic review. \u003cem\u003eInternational Journal of Human Resource Management\u003c/em\u003e, 33(6), 1237\u0026ndash;1266.\u003c/li\u003e\n\u003cli\u003eVrontis, D., Thrassou, A., \u0026amp; Amirkhanpour, M. (2022). B2B technology marketing: The role of AI in decision-making processes. \u003cem\u003eIndustrial Marketing Management\u003c/em\u003e, 101, 123-135.\u003c/li\u003e\n\u003cli\u003eWaldkirch, M., Bucher, E., Schou, P. K., \u0026amp; Gruenwald, E. (2021). Controlled by the algorithm, coached by the crowd\u0026mdash;How HRM activities take shape on digital work platforms in the gig economy. \u003cem\u003eInternational Journal of Human Resource Management\u003c/em\u003e, 32(12), 2643\u0026ndash;2682.\u003c/li\u003e\n\u003cli\u003eWirtz, J., Patterson, P. G., Kunz, W. H., Gruber, T., Lu, V. N., Paluch, S., \u0026amp; Martins, A. (2018). \u0026quot;Brave new world: service robots in the frontline.\u0026quot; \u003cem\u003eJournal of Service Management\u003c/em\u003e, 29(5), 907-931.\u003c/li\u003e\n\u003cli\u003eWyssenbach, T., Zeballos, M., Loosli, S., \u0026amp; Schwaninger, A. (2022). Nonverbal behavior of interviewers influences the competence ratings of observers in recruitment interviews: a study investigating social influence using 360-degree videos with virtual reality and 2D screen displays. \u003cem\u003eVirtual Reality\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(2), 669-686.\u003c/li\u003e\n\u003cli\u003eZhou, Y., Yang, C., Liu, Z., \u0026amp; Gong, L. (2023). Digital technology adoption and innovation performance: A moderated mediation model. \u003cem\u003eTechnology Analysis \u0026amp; Strategic Management\u003c/em\u003e, 1-16.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"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, Decision-Making, AI-Human Collaboration","lastPublishedDoi":"10.21203/rs.3.rs-6885768/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6885768/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe interplay between humans and artificial intelligence (AI) in decision-making has become increasingly intricate and significant. Despite rapid advancements, the literature remains fragmented, with limited integrative frameworks to explain how AI-human dynamics and decision-making typologies shape outcomes. This study addresses this critical gap by conducting a systematic review and bibliometric analysis of 1,004 articles, culminating in a novel conceptual framework. The framework identifies two critical dimensions, AI-human dynamics and decision typologies, that shape decision outcomes and introduces four distinct paradigms of AI-human collaborative decision-making: autonomous execution, guided resolution, collaborative exploration, and augmented discovery. By synthesizing these paradigms, this research advances the theoretical understanding of hybrid decision-making systems and provides actionable insights for organizations navigating complex and AI-driven environments. By elucidating the mechanisms and trade-offs inherent in AI-human collaboration, this work lays a robust foundation for future research on adaptive decision systems in an era marked by accelerating technological change.\u003c/p\u003e","manuscriptTitle":"Advancing Decision-Making through AI-Human Collaboration: A Systematic Review and Conceptual Framework","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-18 10:39:14","doi":"10.21203/rs.3.rs-6885768/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":"f34195c7-ab5c-47e5-b805-40e3b1d75205","owner":[],"postedDate":"August 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-07T16:10:46+00:00","versionOfRecord":{"articleIdentity":"rs-6885768","link":"https://doi.org/10.1007/s10726-026-09980-1","journal":{"identity":"group-decision-and-negotiation","isVorOnly":false,"title":"Group Decision and Negotiation"},"publishedOn":"2026-04-03 15:59:11","publishedOnDateReadable":"April 3rd, 2026"},"versionCreatedAt":"2025-08-18 10:39:14","video":"","vorDoi":"10.1007/s10726-026-09980-1","vorDoiUrl":"https://doi.org/10.1007/s10726-026-09980-1","workflowStages":[]},"version":"v1","identity":"rs-6885768","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6885768","identity":"rs-6885768","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}