Artificial intelligence (AI) is increasingly becoming part of how organisations create, access, share, store and apply knowledge. AI-based tools can support employees and organisations in finding information, generating new knowledge, making knowledge available to others and supporting decision-making. They may also support organisational learning, help retain and reuse existing knowledge, and enable knowledge to be combined and applied in new ways. At the same time, the growing use of AI may create new knowledge risks or change the way existing knowledge risks arise and develop within organisations.

Knowledge risks refer to the likelihood and severity of impacts arising from knowledge-based activities that may affect organisational functioning at any level (Durst and Zieba, 2026). They may originate inside or outside the organisation, involve human, technological and operational dimensions, and produce positive or negative consequences. Human knowledge risks concern employees’ expertise and knowledge behaviours. Technological knowledge risks concern the reliability of and dependency on AI systems. Operational knowledge risks concern how knowledge is retained, shared, protected and applied. Across these dimensions, AI may create opportunities by supporting knowledge preservation, access and use. It may also support knowledge creation, facilitate knowledge sharing across organisational boundaries and make expertise more accessible to employees. It may also create or amplify challenges. For example, organisations may face knowledge loss when employees increasingly rely on AI systems to perform knowledge-intensive tasks. AI may also affect knowledge sharing and knowledge hiding among employees, while the increased use of external AI systems may raise concerns about knowledge leakage and knowledge ownership. Other challenges concern misinformation, the reliability of AI-generated knowledge, the loss of critical expertise and increasing dependency on AI systems and the knowledge embedded in them.

Dealing with these knowledge risks is not necessarily straightforward. Knowledge risks are often connected to one another and may occur at the individual, team, organisational or inter-organisational level. A risk emerging at one level may contribute to the development of another risk elsewhere in the organisation. Their occurrence may also depend on technological, individual, organisational and institutional conditions. Consequently, understanding knowledge risks in the AI era requires not only identifying individual risks but also examining how they are related, under which conditions they occur and how they may develop over time. From an ecology of knowledge perspective, these relationships form part of a wider knowledge ecosystem in which human expertise, organisational routines, relationships, technologies and AI systems interact in the creation, sharing and use of knowledge. Changes in one part of this ecosystem may therefore affect other parts and shape the opportunities and knowledge risks that arise within organisations. A risk that disturbs one part of this ecosystem, such as a loss of tacit expertise or a narrowing of the sources from which knowledge is drawn, may unsettle the diversity and balance of which the ecosystem’s health depends.

This also raises methodological questions. Existing research methods provide different possibilities for studying knowledge risks, but no single method may be suitable for addressing all aspects of these phenomena. Some methods may be useful for examining relationships between variables, while others can help identify combinations of conditions, analyse interdependencies between risks, explore causal structures or capture the views and experiences of organisational actors. Longitudinal and process-based approaches may examine how knowledge risks emerge, interact and develop over time. Multilevel methods may explain how risks originating at one level create or intensify risks at another. Qualitative approaches may examine how organisational actors recognise, interpret and respond to knowledge risks, evaluate AI-generated knowledge, assign responsibility and balance knowledge access with knowledge protection.

Against this background, the organisers of this track encourage academics and practitioners to share their findings and insights on how knowledge risks associated with AI can be identified, analysed, explained and managed. Particular attention is given to the methodological approaches that can help us better understand the complexity and interconnectedness of knowledge risks in AI-enabled organisations.

The track welcomes studies using established methods as well as studies developing, extending or combining methodological approaches. These may include structural equation modelling, configurational approaches such as fsQCA, interpretive structural modelling and its extensions, multi-criteria decision-making methods, social network analysis, experiments, longitudinal research, expert-based approaches, mixed methods, qualitative comparative approaches and computational methods. Qualitative and process-based studies may also use case studies, interviews, ethnography, critical incident studies, process tracing, action research and participatory research. Studies combining different methods to examine the same knowledge-risk phenomenon are also encouraged.

In addition, the track welcomes contributions addressing methodological issues related to the measurement of knowledge risks, construct development, expert judgement, validation, robustness and sensitivity analysis, reproducibility, comparison across organisational or national contexts, and the integration of qualitative and quantitative evidence.

The intention of the track is therefore not to favour a particular methodological approach, but to provide a platform for discussing how different methodological choices can contribute to a better understanding of knowledge risks within the ecology of knowledge in the AI era and support organisations in dealing with these risks.