The Landscape of Generative AI in Information Systems: A Synthesis of Secondary Reviews and Research Agendas

As organizations grapple with the rapid adoption of Generative AI (GenAI), this study synthesizes the state of knowledge through a systematic literature review of secondary studies and research agendas. Analyzing 28 papers published since 2023, we find that while GenAI offers transformative potential for productivity and innovation, its adoption is constrained by multiple interrelated challenges, including technical unreliability (hallucinations, performance drift), societal-ethical risks (bias, misuse, skill erosion), and a systemic governance vacuum (privacy, accountability, intellectual property). Interpreted through a socio-technical lens, these findings reveal a persistent misalignment between GenAI’s fast-evolving technical subsystem and the slower-adapting social subsystem, positioning IS research as critical for achieving joint optimization. To bridge this gap, we discuss a research agenda that reorients IS scholarship from analyzing impacts toward actively shaping the co-evolution of technical capabilities with organizational procedures, societal values, and regulatory institutions–emphasizing hybrid human–AI ensembles, situated validation, design principles for probabilistic systems, and adaptive governance.

💡 Research Summary

This paper conducts a systematic literature review of secondary studies and research‑agenda papers on generative artificial intelligence (GenAI) within the information systems (IS) discipline, covering 28 publications released since 2023. The authors set three objectives: (O1) map the emerging landscape of GenAI‑related secondary literature; (O2) synthesize the benefits and challenges reported across these works; and (O3) derive an integrated, theory‑grounded research agenda for IS scholars.

The review follows a transparent search‑and‑selection protocol, applying inclusion/exclusion criteria and quality appraisal to isolate a coherent sample of reviews, scoping studies, mapping studies, and forward‑looking roadmaps. Bibliometric analysis reveals a rapid proliferation of GenAI literature, with clusters focusing on specific domains (healthcare, education, software engineering), risk dimensions (bias, privacy, reliability), or methodological perspectives.

Synthesizing the findings, the authors identify three overarching benefit categories: (1) productivity gains through automation of routine knowledge work; (2) personalization and innovation enabled by large language models (LLMs) and multimodal generators; and (3) new forms of decision support that blur the line between human judgment and algorithmic suggestion. Conversely, they uncover three interrelated challenge clusters:

• Technical unreliability – hallucinations, performance drift, and model brittleness when deployed outside training distributions.
• Societal‑ethical risks – algorithmic bias, potential misuse, erosion of professional skills, and threats to academic integrity.
• Governance vacuum – insufficient privacy safeguards, unclear accountability, and unresolved intellectual‑property regimes for AI‑generated content.

Interpreting these results through a socio‑technical lens, the paper argues that GenAI’s rapid technical evolution outpaces the slower adaptation of social, organizational, and institutional structures, creating a persistent misalignment. This misalignment positions IS research as a crucial mediator for “joint optimization” of the technical and social subsystems.

To address the gap, the authors propose a four‑pillar research agenda:

1. Hybrid Human‑AI Ensembles – design and evaluate collaborative work structures where humans and GenAI agents complement each other, focusing on role allocation, trust calibration, and skill development.
2. Situated Validation – develop continuous, context‑aware evaluation frameworks that monitor model outputs in real‑world settings, detect drift, and trigger human oversight when uncertainty exceeds thresholds.
3. Design Principles for Probabilistic Systems – formulate guidelines for representing uncertainty, visualizing confidence scores, and communicating the probabilistic nature of generative outputs to end‑users.
4. Adaptive Governance – co‑create policy‑technology‑organization feedback loops that enable dynamic regulation, accountability mechanisms, and IP frameworks responsive to the evolving capabilities of GenAI.

The paper concludes by acknowledging limitations (e.g., the nascent nature of the literature, potential overlap among reviews) and calling for longitudinal, cross‑disciplinary studies that integrate technical performance metrics with social impact assessments. Overall, it offers IS scholars a consolidated map of the current GenAI knowledge base, a balanced view of its promises and perils, and a concrete set of research directions aimed at steering the responsible evolution of generative AI within organizations and society.