This study evaluates the impact of the Consumer Review Fairness Act (CRFA), a federal U.S. law designed to prevent businesses from using legal threats to suppress negative online feedback. Employing a difference-in-differences design on data from TripAdvisor and Google Places, the authors examine shifts in review valence, sentiment, and length following the regulation. Problem Businesses increasingly suppress negative online reviews through legal threats and non-disparagement clauses, severely biasing consumer perceptions and creating a chilling effect among reviewers. Despite increasing policy responses, empirical evidence on whether regulations effectively curb review suppression and restore rating authenticity has been lacking. Outcome - Following the CRFA enactment, U.S. hotel reviews became systematically more negative in both star ratings and textual sentiment.
- Reviews became significantly longer on average, indicating that consumers shared more detailed feedback once legal risks were mitigated.
- Rating declines were most concentrated among less reputable hotels and those in highly competitive local markets, which previously suppressed negative reviews more aggressively.
- The policy's effects were strongest among American reviewers and users with longer platform tenure.

This study explores how integrating generative artificial intelligence (GAI) tools like ChatGPT into higher education coursework can mediate learning and support student literacy. Grounded in technology-mediated learning theory, the authors conduct a qualitative comparative case study analyzing 45 student groups across two business courses in a French business school. Problem Despite widespread recognition of GAI's transformative potential in education, little empirical research exists on how GAI actively functions as an interactive participant in learning processes. Higher education educators lack clear guidance on how to design assignments that effectively develop students' GAI literacy while balancing technological reliance with critical human oversight. Outcome - GAI tools act as active learning mediators that help bridge knowledge disparities and deliver adaptive, scalable support to diverse student groups.
- Experiential engagement with GAI develops crucial literacy skills including prompt engineering, task allocation, and critical output evaluation.
- Human educators remain essential to supply emotional, ethical, and contextual guidance that GAI cannot replicate.

This study investigates how the introduction of a digital platform drives business model change in small- and medium-sized enterprises (SMEs). Using a longitudinal case study of four German textile firms participating in a joint platform initiative, the authors examine the interplay between organizational capabilities and dynamic platform orientations across three implementation phases. Problem While digital platforms provide opportunities for resource-constrained SMEs to collaborate and scale, platform introduction is resource-intensive and frequently fails to produce transformative outcomes. Existing research fails to clarify why similar SMEs sharing the same technological infrastructure and network experience divergent business model outcomes. Outcome - All SMEs initially adopt an efficiency-focused platform orientation as a low-risk, resource-conserving entry logic to streamline operations.
- Deep operational embedding of the platform during the efficiency phase strengthens SMEs' digital, agile, and network capabilities.
- Accumulated capabilities allow firms to dynamically shift toward an innovation-focused orientation, enabling transformative business model changes such as co-branding and forming virtual companies.
- SMEs with shallow platform engagement remain locked in an efficiency orientation, achieving only incremental business model improvements.

This study explores how an immersive metaverse environment can facilitate dynamic collaboration and creative ideation among geographically distributed teams. Using an action research methodology, the researchers designed and evaluated a three-day virtual design-thinking workshop involving 26 software professionals from a digital transformation company. Problem Distributed teams frequently struggle to replicate the spontaneous interactions, creative energy, and psychological safety of physical co-located innovation workshops. Traditional video conferencing platforms often result in low engagement, communication silos, and hesitation to share unpolished or unconventional ideas. Outcome - Immersive avatar-based interactions and informal spatial movement lowered social pressure, significantly enhancing psychological safety and participant engagement.
- Combining spatial metaverse platforms with digital whiteboard tools allowed teams to transition fluidly between open exploration and structured prototyping.
- Skilled, continuous facilitation proved essential for guiding virtual transitions, maintaining momentum, and reducing technical adoption barriers.
- Assembling cross-functional teams across different geographic locations successfully broke organizational silos and fostered more diverse, novel solution concepts.

Drawing on production theory and a stochastic frontier analysis, this study examines how cloud computing affects users' energy efficiency using U.S. industry-level data from 1997 to 2017 alongside a firm-level survey. It distinguishes between software-as-a-service (SaaS) and infrastructure-as-a-service (IaaS) to evaluate their distinct mechanisms and economic energy savings. Problem While cloud computing has seen rapid adoption, intense debate persists regarding its environmental footprint, particularly concerning high energy consumption by vendor data centers. However, little empirical research exists evaluating whether user-side energy efficiency gains offset these vendor-side environmental impacts. Outcome - Cloud-based IT services significantly improve client energy efficiency, resulting in estimated user-side cost savings of $2.8–$12.6 billion in 2017 alone.
- Software-as-a-Service (SaaS) enhances both electric and nonelectric energy efficiency across all industries by driving operational efficiency and reducing resource waste.
- Infrastructure-as-a-Service (IaaS) primarily improves electric energy efficiency in hardware-intensive industries by replacing inefficient internal IT equipment and infrastructure.
- Net energy savings on the user side far outweigh the energy consumed by cloud vendor industries and are comparable to total electricity consumption by U.S. data centers.

This study evaluates whether machine learning models using facial features to estimate learners' mental states (engagement and help-seeking) can generalize across different cultures. Researchers tested Taiwanese participants using an intelligent tutoring system and compared their results to Japanese participants using LightGBM models and SHAP feature analysis. The study aims to clarify cultural similarities and differences in facial action units during online learning. Problem Although facial expressions are effective for tracking student engagement and help-seeking in e-learning contexts, most predictive models are created within a single culture. Because facial behaviors vary culturally, it remains unclear whether these machine learning models can generalize to international learners without performance degradation or bias. Addressing this gap is essential for developing adaptive, cross-cultural intelligent tutoring systems. Outcome - LightGBM models effectively classified engagement and help-seeking states for Taiwanese learners, though accuracy scores were significantly higher for Japanese participants.
- Estimating engagement relied more heavily on facial features around the eyes for Japanese participants, whereas features around the mouth were more important for Taiwanese participants.
- Help-seeking estimations demonstrated significant cultural differences in SHAP values, particularly with mouth-related features like lip stretcher (AU20) and lips part (AU25).
- Cross-cultural correlation analysis revealed that facial expressions of engagement are relatively consistent across cultures, whereas help-seeking expressions are more culturally specific.

This study investigates how computer vision and machine learning techniques can automatically estimate online learners' engagement levels and help-seeking intentions using facial action units and head pose features. Participants performed problem-solving tasks on an intelligent tutoring system prototype while their facial expressions and head movements were recorded and analyzed using LightGBM and Support Vector Machine models. Problem In e-learning environments, instructors struggle to track students' attention and detect when they require assistance due to the absence of non-verbal classroom cues. Existing automated tutoring systems typically fail to identify a student's need for help before an explicit query or error occurs. Outcome - LightGBM classifiers outperformed SVM models, achieving classification accuracy between 0.69 and 0.93 for both engagement and help-seeking states.
- Facial Action Units around the brows and lips (such as outer brow raiser, brow lowerer, lip tightener, and dimpler) proved to be key predictors for learner mental states.
- Head position and orientation features were highly effective indicators, with head pitch being particularly relevant for engagement and head yaw for help-seeking behavior.

This study investigates how acquiring a subscription to a video streaming service like Netflix affects an individual's offline movie theater spending and consumption habits. Using transaction records from a popular mobile wallet app in South Korea, the authors apply coarsened exact matching and difference-in-differences analyses to estimate causal effects. Problem Although digital subscription models are widely believed to cannibalize physical consumption in copyright industries, it remains unclear whether streaming services substitute for offline movie theaters. Because traditional cinema provides distinct social and hedonic value that personal-device streaming cannot fully replicate, there is a need to examine whether these channels act as substitutes or complements. Outcome - Subscribing to Netflix leads to a statistically significant increase in subsequent offline movie ticket spending and viewing frequency.\n- Post-subscription offline moviegoing is frequently accompanied by increased spending on complementary social and hedonic activities such as dining out and entertainment.\n- The positive treatment effect is more pronounced among younger and lower-income individuals who exhibit higher ex-ante social and hedonic needs.\n- Subscribers to premium Netflix plans show a higher post-subscription increase in offline cinema consumption, supporting expectation disconfirmation theory.

This paper presents a generalizable framework for conducting process science studies that leverage digital trace data and computational methods like process mining. Guided by principles of contextual grounding, iterative reasoning, and combining computational with human sense-making, the paper structures process science inquiry into three recursive cycles: descriptive, explanatory, and prescriptive. Problem While process science provides strong conceptual foundations for studying socio-technical processes, concrete methodological guidance on how to conduct empirical process science studies in practice remains underdeveloped. Existing computational techniques excel at revealing what happens in processes, but they lack the contextual and interpretive mechanisms needed to explain why processes unfold as they do or how to design effective interventions. Outcome - Defines three core methodological principles for process science: contextual grounding, iterative reasoning, and the integration of computational and human sense-making.
- Proposes three interconnected, recursive cycles of inquiry: a descriptive cycle (from scope via data to description), an explanatory cycle (from analysis via pattern to explanation), and a prescriptive cycle (from intervention via evaluation to prescription).
- Demonstrates the practical execution of these cycles through a real-world case study of a financial institution's digital customer onboarding process.
- Provides concrete guidance for researchers and practitioners on combining qualitative insights with computational techniques to move from descriptive analytics to actionable organizational interventions.

This study develops a configurational typology of artificial intelligence governance grounded in cultural-ethical fit to explain variations in regulatory approaches worldwide. Drawing on a systematic literature review of 183 studies and established value theories, the authors synthesize how cultural value orientations interact with ethical priorities. The resulting framework provides a structured foundation for analyzing and designing context-sensitive AI governance regimes. Problem Current AI governance frameworks frequently assume that ethical principles like fairness, transparency, and accountability are universally understood and applied. However, exporting standardized regulatory models across diverse national contexts can create severe misalignment with local socio-cultural values. This misalignment reduces the legitimacy of governance regimes, fosters regulatory arbitrage, and creates compliance friction for multinational organizations. Outcome - Identified eight distinct ideal-type AI governance profiles reflecting specific cultural and ethical alignments: group fairness, individual fairness, market-led growth, value-led growth, human oversight fortress, autonomous growth, individual control, and collective security.
- Demonstrated that global AI governance is structured by two key meta-tensions: social justice versus economic competitiveness, and autonomy versus security and control.
- Formulated three theoretical propositions outlining how cultural alignment determines governance legitimacy, shapes patterns of societal resistance, and guides regulatory evolution over time.
- Offered a diagnostic decision framework for policymakers to adapt international AI guidelines to local socio-cultural contexts without sacrificing core safety requirements.

This study investigates how Generative AI (GenAI) reshapes Data Quality Management (DQM) into a collaborative human-AI process through evolving delegation patterns. Adopting a mixed-method approach, the authors conduct a systematic mapping of 209 DQM solutions alongside semi-structured interviews with 11 domain experts to conceptualize GenAI roles and their governance implications. Problem Traditional rule-based data quality management is resource-intensive, rigid, and struggles to scale alongside growing enterprise data complexity. Furthermore, integrating GenAI without strategic human oversight introduces severe organizational risks, such as model hallucinations, data corruption, privacy breaches, and unclear accountability boundaries. Outcome - Formulated a four-role typology for GenAI in DQM: Translator (natural language to rule logic), Explainer (narrative summaries of anomalies and metadata), Resolver (data correction suggestions), and Integrator (cross-system schema alignment).
- Revealed that current GenAI adoption in commercial DQM tools remains nascent (~17%), functioning largely as an interpretive co-pilot rather than an autonomous decision-maker.
- Proposed five theoretical governance guidelines highlighting task decomposability appraisal, mandatory human validation for high-impact actions, reverse delegation protocols, explainability standards, and retained human accountability.

This study investigates how adopting smart speakers affects consumers' digital content search, purchase, and consumption behaviors. Utilizing household-level panel data on Video on Demand (VOD) services along with remote-control clickstream data and user surveys, the authors apply propensity score matching and difference-in-differences estimation to isolate the economic impacts of voice channels. Problem While voice-assisted technologies are rapidly penetrating households, it remains uncertain how they influence consumer decision-making processes across search, purchase, and consumption stages. Specifically, firms lack empirical insights into how lower search costs in voice channels alter consumer effort, decision friction, and digital content preferences. Outcome - Smart speaker adoption led to a 21.5% increase in monthly VOD content expenditures, alongside a 3.0% decrease in average content completion rates.
- The increase in purchases is driven by lowered search friction, resulting in more search incidences, higher conversion rates, and less labor-intensive evaluation of content.
- Using smart speakers for conversational purposes beyond simple informational queries amplifies the positive impact of device adoption on sales.

This study investigates how organisations can effectively design and deploy artificial intelligence-augmented decision making (AIADM) systems. Using an action design research methodology within an online fashion retail company, the authors developed and evaluated an AIADM system to support decisions across customer segmentation, customer retention, and product portfolio redesign. Problem Traditional decision support systems rely on predefined rules and human agentic primacy, making existing design knowledge difficult to apply directly to adaptive AI systems. Organisations face significant challenges in adopting AIADM systems, including stochastic outputs, ambiguous decision authority, regulatory concerns, and resource constraints. Outcome - Formulate a strategic data roadmap aligning AIADM use cases with business model requirements, resources, and capabilities.
- Ensure synergy across input data, AI models, and visual outputs to demonstrate tangible business value.
- Implement ethical AI governance frameworks and auditing mechanisms to ensure compliance and build stakeholder trust.
- Maintain human involvement in design and execution to integrate domain expertise and retain decision accountability.
- Embrace an iterative design mindset focused on continuous learning, adaptation, and system improvement over time.
- Utilize open-source tools and external expert partnerships to mitigate talent and resource limitations.

This scoping review synthesizes existing research on responsible artificial intelligence (AI) principles and governance practices across the complete AI project lifecycle. The study develops a holistic conceptual framework that details structural, procedural, and relational practices along with their antecedents and organizational and societal effects. Problem Despite widespread guidelines and high-level principles for ethical AI, organizations struggle to translate abstract concepts into actionable practices. Literature in this domain remains disparate and lacks clarity on how to operationalize responsible AI governance while balancing business competitiveness, ethical risks, and compliance requirements. Outcome - Synthesizes prior research on responsible AI into seven core principles: accountability, diversity and fairness, human agency and oversight, privacy and data governance, technical robustness and safety, transparency, and social and environmental well-being.
- Establishes a tripartite framework for responsible AI governance comprising structural (roles and authority), procedural (processes and evaluation), and relational (collaboration and literacy) practices.
- Maps the contextual antecedents of AI governance, including evolving societal norms and organizational values, as well as its effects on business value and social assessment.
- Formulates a detailed research agenda with key questions to guide future academic inquiry and practical implementation.

This study examines the collaborative redesign of an introductory information systems course at a US university through the theoretical lenses of Communities of Practice and Social Cognitive Theory. Using a participatory action research approach, the authors analyzed qualitative interview data, field notes, and student pre- and post-test assessment scores to evaluate the impact on course quality and instructor self-efficacy. Problem Introductory multi-section university courses frequently suffer from inconsistent instructional quality, varying depth of coverage, and minimal collaboration among instructors. Additionally, standard faculty development models often fail to provide experienced instructors with effective collaborative mechanisms that enhance their confidence to adopt shared pedagogical practices. Outcome - Student performance showed consistent, measurable gains across multiple semesters following the course redesign, as demonstrated by improved pre- and post-test assessment deltas.
- Participation in the community of practice positively impacted instructors' self-efficacy beliefs, with vicarious learning and peer verbal persuasion emerging as key drivers of confidence.
- The collaborative approach established high course consistency across sections while creating a durable, supportive framework for peer mentoring and ongoing curriculum enhancement.

This study investigates how information systems (IS) practitioner doctorates bridge the gap between academic research and practical industry application. Utilizing Pierre Bourdieu's Theory of Practice, the authors examine a multi-case study of 10 research projects using data from 24 semi-structured interviews and participant observations. The research explores the boundary spanning activities and environmental forces that enable researchers to achieve practical impact across three stages of development. Problem Demonstrating real-world practical impact alongside theoretical contributions remains a major challenge in information systems research due to the traditional divide between academia and practice. Existing literature lacks empirical insights into how transitional agents, such as practitioner doctorates, navigate cross-field boundaries, overcome institutional norms, and align divergent goals to create actionable outcomes. Outcome - IS practitioner doctorates facilitate knowledge exchange through boundary-spanning activities including research design adaptation, multi-channel dissemination, and creating symbolic discourse.
- Practical impact progresses through proof-of-concept, proof-of-value, and proof-of-use stages, each requiring different combinations of social, cultural, economic, and symbolic capital.
- Environmental forces such as academic vs. practice doxa (institutional norms) and habitus (professional dispositions) create positional conflicts that require active negotiation and tailored incentives to sustain engagement.

This editorial paper introduces and conceptualizes process science as an interdisciplinary field dedicated to the study of socio-technical processes over time. It examines how human actions and digital technologies interact dynamically, leveraging digital trace data and computational techniques to observe, explain, and guide change. The study establishes core tenets, taxonomy distinctions, and a research framework linking practical socio-technical processes with scientific knowledge creation. Problem Traditional research disciplines frequently prioritize an 'entity-first' view, focusing on static structures rather than ongoing, dynamic changes. However, contemporary phenomena—such as generative AI, platform economies, and societal transformations—are inherently socio-technical and continuously evolving, producing massive amounts of digital trace data that static analytical frameworks cannot adequately capture. Outcome - Defines process science as an interdisciplinary domain aimed at capturing, understanding, and intervening in socio-technical change across multiple levels of abstraction.
- Categorizes key activities of process science into discovery (detecting patterns via event data), explanation (identifying cause-effect relations), and intervention (shaping process trajectories).
- Advocates for shifting from an 'entity-first' to a 'process-first' perspective to identify analogies across diverse domains and support computationally intensive theory building.
- Introduces a research framework that integrates digital trace data from practice with theoretical foundations to solve real-world problems and drive societal impact.

This study investigates how individuals with lifelong disabilities utilize virtual worlds, specifically Second Life, to move from social isolation to social inclusion. Through a qualitative research design involving participant observation and in-depth interviews, the authors trace how users perceive and actualize virtual world capabilities. The paper develops an affordance-based pathway model outlining how technology features enable interpersonal relationships and community participation. Problem People with lifelong disabilities frequently face severe social exclusion, isolation, and physical barriers that limit their full participation in everyday social activities. While digital environments offer potential solutions, the underlying mechanisms and technological pathways through which virtual tools lead to meaningful social inclusion remain insufficiently understood. Outcome - Identified three functional affordances (communicability, mobility, and personalizability) and two social affordances (engageability and self-actualizability) crucial for social inclusion.
- Established that the actualization of functional affordances enables and dynamically interacts with social affordances to foster interpersonal relationships and community participation.
- Discovered that outcomes achieved by actualizing specific affordances serve as facilitating conditions that support the perception and realization of additional affordances.
- Extended affordance theory by demonstrating a feedback loop where social engagement refines and enhances the perception and use of basic functional capabilities.