Pub Date : 2025-11-28DOI: 10.1016/j.ijinfomgt.2025.103006
Zhengfu Wang, Weiwei Wu
This paper examines the impact of public data openness on firm narrative R&D disclosure by leveraging the launch of public data platforms as a policy shift. Using a Difference-in-Differences (DID) approach, our findings reveal that firms significantly reduce their narrative R&D disclosures following the implementation of public data openness. This effect is stronger for firms with higher R&D intensity and those operating in more competitive industries. Our study contributes to the literature on R&D information flows by highlighting the unintended consequences of public data openness. We also discuss practical recommendations to mitigate the potential negative effects on the R&D information dissemination.
{"title":"The impact of public data openness on firm narrative R&D disclosure","authors":"Zhengfu Wang, Weiwei Wu","doi":"10.1016/j.ijinfomgt.2025.103006","DOIUrl":"10.1016/j.ijinfomgt.2025.103006","url":null,"abstract":"<div><div>This paper examines the impact of public data openness on firm narrative R&D disclosure by leveraging the launch of public data platforms as a policy shift. Using a Difference-in-Differences (DID) approach, our findings reveal that firms significantly reduce their narrative R&D disclosures following the implementation of public data openness. This effect is stronger for firms with higher R&D intensity and those operating in more competitive industries. Our study contributes to the literature on R&D information flows by highlighting the unintended consequences of public data openness. We also discuss practical recommendations to mitigate the potential negative effects on the R&D information dissemination.</div></div>","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"87 ","pages":"Article 103006"},"PeriodicalIF":27.0,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145616133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-26DOI: 10.1016/j.ijinfomgt.2025.103004
Shameem Shagirbasha , Naman Agarwal , Angelin Vilma G.
In a labor-intensive sector such as healthcare, the work productivity of frontline healthcare workers (FHWs) is crucial to reducing costs and managing patient volume. This study explores the affordances of Gen AI HITs that enhance FHWs’ work productivity and examines the mechanisms underlying this effect. A sequential mixed-methods design was employed for this study: qualitative interviews with 32 FHWs to identify the affordances that positively influence work productivity, followed by quantitative analyses using the PROCESS macro and structural equation modeling (SEM) to assess mediation by techno-eustress and moderation by job self-efficacy. The qualitative findings indicate that Gen AI HITs’ information, navigation, and interactivity affordances foster work productivity among FHWs, among other affordances identified. The quantitative results highlight that techno-eustress mediates the positive impact of Gen AI HITs’ interactivity and information affordances on FHWs’ work productivity, but not navigation affordance. However, when accounting for FHWs’ job self-efficacy, the mediation effect of techno-eustress becomes significant for all three affordances of Gen AI HIT – information, navigation, and interactivity. Specifically, the indirect positive impact of these affordances on productivity is stronger among FHWs with higher job self-efficacy. These results offer significant contributions to understanding the human–technology interaction in healthcare and provide practical insights for designing Gen AI HITs and training programs that improve adoption while enhancing work performance.
在医疗保健等劳动密集型行业，一线医疗工作者（FHWs）的工作效率对于降低成本和管理患者数量至关重要。本研究探讨了新一代人工智能HITs在提高fhw工作效率方面的优势，并研究了这种影响的潜在机制。本研究采用顺序混合方法设计：对32名外籍家庭佣工进行定性访谈，以确定对工作效率产生积极影响的支持，然后使用PROCESS宏观和结构方程模型（SEM）进行定量分析，以评估技术压力的中介作用和工作自我效能的调节作用。定性研究结果表明，Gen AI HITs的信息、导航和交互性能力提高了fhw的工作效率，以及其他已确定的能力。定量结果强调，技术压力介导了Gen AI HITs的交互性和信息能力对FHWs工作效率的积极影响，但不影响导航能力。然而，当考虑到FHWs的工作自我效能感时，技术压力对Gen AI HIT的信息、导航和交互性三种能力的中介作用都是显著的。具体而言，这些能力支持对工作效率的间接积极影响在工作自我效能感较高的外籍佣工中更为明显。这些结果为理解医疗保健领域的人机交互做出了重大贡献，并为设计Gen AI hit和培训计划提供了实际见解，从而在提高工作绩效的同时提高采用率。
{"title":"Artificial intelligence in healthcare IT: Enhancing work productivity through techno-eustress","authors":"Shameem Shagirbasha , Naman Agarwal , Angelin Vilma G.","doi":"10.1016/j.ijinfomgt.2025.103004","DOIUrl":"10.1016/j.ijinfomgt.2025.103004","url":null,"abstract":"<div><div>In a labor-intensive sector such as healthcare, the work productivity of frontline healthcare workers (FHWs) is crucial to reducing costs and managing patient volume. This study explores the affordances of Gen AI HITs that enhance FHWs’ work productivity and examines the mechanisms underlying this effect. A sequential mixed-methods design was employed for this study: qualitative interviews with 32 FHWs to identify the affordances that positively influence work productivity, followed by quantitative analyses using the PROCESS macro and structural equation modeling (SEM) to assess mediation by techno-eustress and moderation by job self-efficacy. The qualitative findings indicate that Gen AI HITs’ information, navigation, and interactivity affordances foster work productivity among FHWs, among other affordances identified. The quantitative results highlight that techno-eustress mediates the positive impact of Gen AI HITs’ interactivity and information affordances on FHWs’ work productivity, but not navigation affordance. However, when accounting for FHWs’ job self-efficacy, the mediation effect of techno-eustress becomes significant for all three affordances of Gen AI HIT – information, navigation, and interactivity. Specifically, the indirect positive impact of these affordances on productivity is stronger among FHWs with higher job self-efficacy. These results offer significant contributions to understanding the human–technology interaction in healthcare and provide practical insights for designing Gen AI HITs and training programs that improve adoption while enhancing work performance.</div></div>","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"87 ","pages":"Article 103004"},"PeriodicalIF":27.0,"publicationDate":"2025-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145616132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-26DOI: 10.1016/j.ijinfomgt.2025.103010
Hui Yang , Yu Zeng , Huizi Xing , Peng Hu
Generative AI (GenAI) systems like ChatGPT offer immense potential but also introduce unique challenges, particularly for users navigating uncertainty in GenAI interactions. This study focuses on two distinct uncertainties: prompt uncertainty (uncertainty about how to phrase effective prompts) and response uncertainty (uncertainty about how GenAI will respond even for the same prompt). We examine how these uncertainties contribute to user fatigue and influence feedback behavior. Using data collected from 832 GenAI users, we find that prompt uncertainty induces emotional fatigue, whereas response uncertainty triggers cognitive fatigue. Furthermore, both types of fatigue can reduce users' willingness to provide feedback to GenAI (e.g., rating GenAI outputs or reporting GenAI errors), which can hinder the iterative refinement of GenAI performance. By disentangling the distinct impacts of these uncertainties, this study contributes to a deeper understanding of GenAI-induced fatigue and its implications for user behavior. The findings also offer insights for GenAI developers to address uncertainty and mitigate user fatigue, ultimately fostering sustained user engagement and improving feedback mechanisms.
{"title":"Fatigued by uncertainties: Exploring the cognitive and emotional costs of generative AI usage","authors":"Hui Yang , Yu Zeng , Huizi Xing , Peng Hu","doi":"10.1016/j.ijinfomgt.2025.103010","DOIUrl":"10.1016/j.ijinfomgt.2025.103010","url":null,"abstract":"<div><div>Generative AI (GenAI) systems like ChatGPT offer immense potential but also introduce unique challenges, particularly for users navigating uncertainty in GenAI interactions. This study focuses on two distinct uncertainties: prompt uncertainty (uncertainty about how to phrase effective prompts) and response uncertainty (uncertainty about how GenAI will respond even for the same prompt). We examine how these uncertainties contribute to user fatigue and influence feedback behavior. Using data collected from 832 GenAI users, we find that prompt uncertainty induces emotional fatigue, whereas response uncertainty triggers cognitive fatigue. Furthermore, both types of fatigue can reduce users' willingness to provide feedback to GenAI (e.g., rating GenAI outputs or reporting GenAI errors), which can hinder the iterative refinement of GenAI performance. By disentangling the distinct impacts of these uncertainties, this study contributes to a deeper understanding of GenAI-induced fatigue and its implications for user behavior. The findings also offer insights for GenAI developers to address uncertainty and mitigate user fatigue, ultimately fostering sustained user engagement and improving feedback mechanisms.</div></div>","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"87 ","pages":"Article 103010"},"PeriodicalIF":27.0,"publicationDate":"2025-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145616135","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-21DOI: 10.1016/j.ijinfomgt.2025.103003
Xuefei (Nancy) Deng , Rui Sun
Artificial intelligence (AI) is disrupting workforce and posing an unprecedented threat of job displacement. However, our understanding of AI's role in shaping individual career development is limited. This study provides insights into AI and career development within the context of first-generation college students (FGCSs), a marginalized group that is arguably among the most vulnerable to the career disruption of AI. Employing mixed methods, this exploratory study examines the effects of FGCS status and career anchor on individual concerns about AI’s career impact and the perceptions of FGCSs and non-FGCSs regarding their career development. Using survey data from 70 students at a minority-serving public university in the United States, the quantitative analysis shows that FGCS status is positively associated with individual concern about AI’s career impact, whereas prior ChatGPT experience is negatively associated with this concern. However, we did not find evidence that a student’s career anchor affects their concerns about AI’s career impact. Meanwhile, the qualitative analysis revealed four themes that highlight employed FGCSs’ reliance on college education to change to a professional career or prepare for entrepreneurship. Our follow-up study revealed four types of individual attitudes toward AI’s career impact and suggested that the attitudes are influenced by generational status and career stage. We compare FGCSs and their peers in terms of career stage, career development and attitude toward AI’s impact and propose intervention strategies to help FGCSs mitigate AI-related job replacement risks. The study contributes to research on the AI impact on career development of a marginalized population.
{"title":"Artificial intelligence and career development: Concerns and insights from first-generation college students","authors":"Xuefei (Nancy) Deng , Rui Sun","doi":"10.1016/j.ijinfomgt.2025.103003","DOIUrl":"10.1016/j.ijinfomgt.2025.103003","url":null,"abstract":"<div><div>Artificial intelligence (AI) is disrupting workforce and posing an unprecedented threat of job displacement. However, our understanding of AI's role in shaping individual career development is limited. This study provides insights into AI and career development within the context of first-generation college students (FGCSs), a marginalized group that is arguably among the most vulnerable to the career disruption of AI. Employing mixed methods, this exploratory study examines the effects of FGCS status and career anchor on individual concerns about AI’s career impact and the perceptions of FGCSs and non-FGCSs regarding their career development. Using survey data from 70 students at a minority-serving public university in the United States, the quantitative analysis shows that FGCS status is positively associated with individual concern about AI’s career impact, whereas prior ChatGPT experience is negatively associated with this concern. However, we did not find evidence that a student’s career anchor affects their concerns about AI’s career impact. Meanwhile, the qualitative analysis revealed four themes that highlight employed FGCSs’ reliance on college education to change to a professional career or prepare for entrepreneurship. Our follow-up study revealed four types of individual attitudes toward AI’s career impact and suggested that the attitudes are influenced by generational status and career stage. We compare FGCSs and their peers in terms of career stage, career development and attitude toward AI’s impact and propose intervention strategies to help FGCSs mitigate AI-related job replacement risks. The study contributes to research on the AI impact on career development of a marginalized population.</div></div>","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"87 ","pages":"Article 103003"},"PeriodicalIF":27.0,"publicationDate":"2025-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145571058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.ijinfomgt.2025.103001
Su Cui , Longdong Wang , Weihang Cao , Tongqing Zhu
The distinct advantages of artificial intelligence (AI) in cognitive and creative projects have driven organizations to advocate for and implement AI, which has contributed to a deep and widespread dependence on AI among employees in creative generation. However, why, how, and when the dependence on AI influences employees’ creativity remains understudied. To figure out these issues, this research explored the double-edged effect of employee dependence on AI on their creativity, drawing on the job demands-resources model. Our mixed methods reveal that employee dependence on AI positively and indirectly affects their creativity via creative process engagement, while the indirect effect is stronger when employee cognitive flexibility is higher than lower. In contrast, employee dependence on AI negatively affects their creativity via information overload when cognitive flexibility is low. These findings have several theoretical and managerial implications related to AI-creativity research and practice.
{"title":"Gain or loss? The dual effects of dependence on AI on employee’s creativity","authors":"Su Cui , Longdong Wang , Weihang Cao , Tongqing Zhu","doi":"10.1016/j.ijinfomgt.2025.103001","DOIUrl":"10.1016/j.ijinfomgt.2025.103001","url":null,"abstract":"<div><div>The distinct advantages of artificial intelligence (AI) in cognitive and creative projects have driven organizations to advocate for and implement AI, which has contributed to a deep and widespread dependence on AI among employees in creative generation. However, why, how, and when the dependence on AI influences employees’ creativity remains understudied. To figure out these issues, this research explored the double-edged effect of employee dependence on AI on their creativity, drawing on the job demands-resources model. Our mixed methods reveal that employee dependence on AI positively and indirectly affects their creativity via creative process engagement, while the indirect effect is stronger when employee cognitive flexibility is higher than lower. In contrast, employee dependence on AI negatively affects their creativity via information overload when cognitive flexibility is low. These findings have several theoretical and managerial implications related to AI-creativity research and practice.</div></div>","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"87 ","pages":"Article 103001"},"PeriodicalIF":27.0,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145571059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.ijinfomgt.2025.103002
Bo Yang, Yongqiang Sun, Zihan Zeng, Qinwei Li
The proliferation of generative AI (GAI) like ChatGPT is transforming how students engage with information and knowledge-focused activities in higher education, sparking debate about its dual impact on learning. While GAI offers potential benefits like enhanced efficiency, concerns about risks such as skill erosion persist. To address this tension, we investigate how students’ dependence on GAI shapes their learning outcomes through skill adaptation processes and under what conditions these effects occur. We conducted a three-phase mixed-methods study (survey N = 306; interviews N = 16; experiment N = 397). Our findings reveal that GAI dependence, influenced by individual learning goals (performance-avoidance/-approach), drives three distinct skill adaptation processes: deskilling (skill erosion), reskilling (acquiring new GAI-related competencies), and upskilling (enhancing existing skills). These adaptations, in turn, differentially impact routine and innovative performance. Qualitative results corroborate and complement these findings, indicating that task characteristics shape GAI use patterns into substitutive and augmentative use. Finally, a scenario-based experiment provides causal evidence for this emergent insight, demonstrating how task characteristics drive the adoption of substitutive vs. augmentative use, which in turn leads to divergent skill adaptation pathways. By combining diverse methodologies, this study clarifies the lights and shadows of GAI dependence, demonstrating how its effects are contingent on individual agency, technological appropriation (substitutive vs. augmentative), and task context. Our findings advance theory on human-AI adaptation and provide practical guidance for practitioners to optimize GAI’s role in learning and knowledge-focused activities.
{"title":"Deskilling, reskilling, or upskilling? Unpacking the pathways of student adaptation to generative artificial intelligence","authors":"Bo Yang, Yongqiang Sun, Zihan Zeng, Qinwei Li","doi":"10.1016/j.ijinfomgt.2025.103002","DOIUrl":"10.1016/j.ijinfomgt.2025.103002","url":null,"abstract":"<div><div>The proliferation of generative AI (GAI) like ChatGPT is transforming how students engage with information and knowledge-focused activities in higher education, sparking debate about its dual impact on learning. While GAI offers potential benefits like enhanced efficiency, concerns about risks such as skill erosion persist. To address this tension, we investigate how students’ dependence on GAI shapes their learning outcomes through skill adaptation processes and under what conditions these effects occur. We conducted a three-phase mixed-methods study (survey N = 306; interviews N = 16; experiment N = 397). Our findings reveal that GAI dependence, influenced by individual learning goals (performance-avoidance/-approach), drives three distinct skill adaptation processes: deskilling (skill erosion), reskilling (acquiring new GAI-related competencies), and upskilling (enhancing existing skills). These adaptations, in turn, differentially impact routine and innovative performance. Qualitative results corroborate and complement these findings, indicating that task characteristics shape GAI use patterns into substitutive and augmentative use. Finally, a scenario-based experiment provides causal evidence for this emergent insight, demonstrating how task characteristics drive the adoption of substitutive vs. augmentative use, which in turn leads to divergent skill adaptation pathways. By combining diverse methodologies, this study clarifies the lights and shadows of GAI dependence, demonstrating how its effects are contingent on individual agency, technological appropriation (substitutive vs. augmentative), and task context. Our findings advance theory on human-AI adaptation and provide practical guidance for practitioners to optimize GAI’s role in learning and knowledge-focused activities.</div></div>","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"87 ","pages":"Article 103002"},"PeriodicalIF":27.0,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145537072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-15DOI: 10.1016/j.ijinfomgt.2025.103000
Michael Chau, Andrew Schwarz
{"title":"Building on the legacy of the International Journal of Information Management","authors":"Michael Chau, Andrew Schwarz","doi":"10.1016/j.ijinfomgt.2025.103000","DOIUrl":"10.1016/j.ijinfomgt.2025.103000","url":null,"abstract":"","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"86 ","pages":"Article 103000"},"PeriodicalIF":27.0,"publicationDate":"2025-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145693140","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-11DOI: 10.1016/j.ijinfomgt.2025.102995
Yujing Xu , Wen-Lung Shiau
Moderation analysis is a critical part in business and management research, particularly within the information systems (IS) domain, yet it continues to face persistent methodological issues. These issues not only threaten the reliability of results but also hinder theoretical advancements. To address these challenges, our study undertakes a comprehensive examination of moderation analysis. We commence with a concise synthesis of its conceptual evolution by reviewing 30 foundational publications that have shaped its development. Subsequently, we categorize and clarify the core moderation models, including two-way, three-way interactions, and moderated mediation, highlighting their appropriate application contexts and corresponding analytical techniques. Building upon this foundational knowledge, we identify and detail 15 prevalent methodological issues in moderation research, assessing their contemporary prevalence through an empirical investigation of 274 articles published in top-tier IS journals over the past three years. To equip researchers with actionable guidance, we propose a state-of-the-art, stage-based framework that encompasses the entire research lifecycle—from initial preparation and hypothesis development through design planning, data collection, sophisticated analysis, rigorous interpretation, to transparent reporting. Our contributions are fourfold. First, we present contemporary empirical evidence on the persistence of historical issues and identify emerging trends in moderation research. Second, we offer a comprehensive, stage-based framework that transcends existing piecemeal recommendations, providing actionable support across the research lifecycle. Third, we consolidate theoretical insights by tracing the conceptual evolution of moderation analysis and systematically classifying major moderation models. Finally, we address the identified critical issues throughout the research process, equipping researchers with empirically validated status assessments and evidence-based solutions. Overall, our study enriches the understanding of moderation analysis and equips researchers, journal editors, and practitioners with a robust methodological roadmap for conducting rigorous and theoretically informed moderation research.
{"title":"Moderation analysis in business and management research: Common issues, solutions, and guidelines for future research","authors":"Yujing Xu , Wen-Lung Shiau","doi":"10.1016/j.ijinfomgt.2025.102995","DOIUrl":"10.1016/j.ijinfomgt.2025.102995","url":null,"abstract":"<div><div>Moderation analysis is a critical part in business and management research, particularly within the information systems (IS) domain, yet it continues to face persistent methodological issues. These issues not only threaten the reliability of results but also hinder theoretical advancements. To address these challenges, our study undertakes a comprehensive examination of moderation analysis. We commence with a concise synthesis of its conceptual evolution by reviewing 30 foundational publications that have shaped its development. Subsequently, we categorize and clarify the core moderation models, including two-way, three-way interactions, and moderated mediation, highlighting their appropriate application contexts and corresponding analytical techniques. Building upon this foundational knowledge, we identify and detail 15 prevalent methodological issues in moderation research, assessing their contemporary prevalence through an empirical investigation of 274 articles published in top-tier IS journals over the past three years. To equip researchers with actionable guidance, we propose a state-of-the-art, stage-based framework that encompasses the entire research lifecycle—from initial preparation and hypothesis development through design planning, data collection, sophisticated analysis, rigorous interpretation, to transparent reporting. Our contributions are fourfold. First, we present contemporary empirical evidence on the persistence of historical issues and identify emerging trends in moderation research. Second, we offer a comprehensive, stage-based framework that transcends existing piecemeal recommendations, providing actionable support across the research lifecycle. Third, we consolidate theoretical insights by tracing the conceptual evolution of moderation analysis and systematically classifying major moderation models. Finally, we address the identified critical issues throughout the research process, equipping researchers with empirically validated status assessments and evidence-based solutions. Overall, our study enriches the understanding of moderation analysis and equips researchers, journal editors, and practitioners with a robust methodological roadmap for conducting rigorous and theoretically informed moderation research.</div></div>","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"86 ","pages":"Article 102995"},"PeriodicalIF":27.0,"publicationDate":"2025-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528685","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-08DOI: 10.1016/j.ijinfomgt.2025.102994
Wansi Chen , Anya Li , Chengkai Pan , Ting Yu , Aamir Ali , Yuanyuan Sun
As algorithmic systems become increasingly embedded in organizational processes, algorithmic management in the workplace has emerged as a central mechanism for guiding, evaluating, and coordinating employee attitudes and behaviors. While existing research has extensively examined the application of algorithmic management in gig platforms, there remains a lack of systematic review and theoretical integration concerning its diverse configurations, deployment conditions, and employee response mechanisms within standardized workplace settings. To address this gap, we conducted a systematic review of 167 peer-reviewed articles on workplace algorithmic management using the BERTopic topic modeling method. Guided by socio-technical systems (STS) theory and a bi-dimensional framework of algorithmic and employee autonomy, we identify four archetypal configurations: surveillance, supervision, supplementary, and complementary. These archetypes reflect distinct employee-algorithm interaction logics across role allocation, task interdependence, and goal alignment. We further examine the technological and organizational conditions required for each configuration and synthesize employee responses across cognitive, emotional, and behavioral domains. By constructing a configuration-based taxonomy rooted in the (in)consistencies of employee-algorithm autonomy, this study explicates the socio-technical deployment mechanisms underlying each archetype and illustrates how employees adapt to algorithmic systems through complex and dynamic engagement trajectories. Our findings offer an integrative framework linking configuration logics, deployment demands, and response patterns, contributing to a more nuanced understanding of how intelligent systems reshape organizational structures and employee experiences.
{"title":"Algorithmic management in the workplace: A systematic review and topic modeling integration using BERTopic","authors":"Wansi Chen , Anya Li , Chengkai Pan , Ting Yu , Aamir Ali , Yuanyuan Sun","doi":"10.1016/j.ijinfomgt.2025.102994","DOIUrl":"10.1016/j.ijinfomgt.2025.102994","url":null,"abstract":"<div><div>As algorithmic systems become increasingly embedded in organizational processes, algorithmic management in the workplace has emerged as a central mechanism for guiding, evaluating, and coordinating employee attitudes and behaviors. While existing research has extensively examined the application of algorithmic management in gig platforms, there remains a lack of systematic review and theoretical integration concerning its diverse configurations, deployment conditions, and employee response mechanisms within standardized workplace settings. To address this gap, we conducted a systematic review of 167 peer-reviewed articles on workplace algorithmic management using the BERTopic topic modeling method. Guided by socio-technical systems (STS) theory and a bi-dimensional framework of algorithmic and employee autonomy, we identify four archetypal configurations: surveillance, supervision, supplementary, and complementary. These archetypes reflect distinct employee-algorithm interaction logics across role allocation, task interdependence, and goal alignment. We further examine the technological and organizational conditions required for each configuration and synthesize employee responses across cognitive, emotional, and behavioral domains. By constructing a configuration-based taxonomy rooted in the (in)consistencies of employee-algorithm autonomy, this study explicates the socio-technical deployment mechanisms underlying each archetype and illustrates how employees adapt to algorithmic systems through complex and dynamic engagement trajectories. Our findings offer an integrative framework linking configuration logics, deployment demands, and response patterns, contributing to a more nuanced understanding of how intelligent systems reshape organizational structures and employee experiences.</div></div>","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"86 ","pages":"Article 102994"},"PeriodicalIF":27.0,"publicationDate":"2025-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145474064","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-07DOI: 10.1016/j.ijinfomgt.2025.102996
Hayeon Kim, Sang Woo Lee
While generative artificial intelligence (AI) has revolutionized various fields, it also presents a significant challenge: 'hallucinations'—plausible but inaccurate information generated by AI systems. Because hallucinations are difficult to prevent entirely, it is essential for generative AI systems to address these inaccuracies effectively. This study investigates how generative AI response strategies to hallucinations affect user satisfaction and tolerance. We examined the impact of politeness (Gratitude vs. Apology) and attribution (Internal vs. External) strategies, as well as AI anthropomorphism, on user reactions. In a 2 × 2 online experiment with 369 ChatGPT users, participants were randomly assigned to one of four response strategy conditions. Results show that users reported the highest satisfaction when the AI apologized and accepted internal responsibility for the error. This effect was particularly pronounced among users who perceived the AI as less human-like, though positive reactions were also observed among users who anthropomorphized the AI. Moreover, user satisfaction mediated the relationship between the AI’s apology/internal attribution and tolerance for hallucinations. This indirect effect was strongest among those who perceived the AI as less human-like. These findings offer theoretical insights into how social response strategies shape user tolerance of AI errors and provide practical guidance for designing more trustworthy and human-centered AI.
{"title":"Sorry, it's my fault: Politeness, attribution, and anthropomorphism in managing generative AI hallucinations","authors":"Hayeon Kim, Sang Woo Lee","doi":"10.1016/j.ijinfomgt.2025.102996","DOIUrl":"10.1016/j.ijinfomgt.2025.102996","url":null,"abstract":"<div><div>While generative artificial intelligence (AI) has revolutionized various fields, it also presents a significant challenge: 'hallucinations'—plausible but inaccurate information generated by AI systems. Because hallucinations are difficult to prevent entirely, it is essential for generative AI systems to address these inaccuracies effectively. This study investigates how generative AI response strategies to hallucinations affect user satisfaction and tolerance. We examined the impact of politeness (Gratitude vs. Apology) and attribution (Internal vs. External) strategies, as well as AI anthropomorphism, on user reactions. In a 2 × 2 online experiment with 369 ChatGPT users, participants were randomly assigned to one of four response strategy conditions. Results show that users reported the highest satisfaction when the AI apologized and accepted internal responsibility for the error. This effect was particularly pronounced among users who perceived the AI as less human-like, though positive reactions were also observed among users who anthropomorphized the AI. Moreover, user satisfaction mediated the relationship between the AI’s apology/internal attribution and tolerance for hallucinations. This indirect effect was strongest among those who perceived the AI as less human-like. These findings offer theoretical insights into how social response strategies shape user tolerance of AI errors and provide practical guidance for designing more trustworthy and human-centered AI.</div></div>","PeriodicalId":48422,"journal":{"name":"International Journal of Information Management","volume":"86 ","pages":"Article 102996"},"PeriodicalIF":27.0,"publicationDate":"2025-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145474062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号