The Theme Development Meetings are planned as annual workshops designed to identify current trends and needs in Alzheimer’s disease research.
These thematic, stakeholder-focused events - held virtually and/or in person - are structured to foster emerging dialogue between the research community and policymakers (e.g. EU officials) across Europe and globally.
On 6 March 2026, the CLARA project held its first Theme Development Meeting with members of the INDRC Scientific Advisory Board (SAB), marking the launch of a new series of expert discussions on trends and priorities in Alzheimer’s disease (AD) research and their alignment with advances in artificial intelligence. The session, titled “AI Architectures for Longitudinal Disease Research: Validation, Governance, and Strategic Positioning”, highlighted two major trends: rapid advances in AI and increasing access to long-term health data, which together create new possibilities for understanding disease progression.
The meeting featured a guest presentation by Dr. Gully Burns titled “Skillful Alhazen: AI Agents for Scientific Knowledge Curation,” which introduced an open-source, agent-based biocuration system designed to support more transparent, structured, and traceable engagement with scientific knowledge. Dr. Burns’ presentation highlighted the limitations of current AI tools, which often produce useful results but offer limited insight into how they are generated, and outlined an approach to make the underlying evidence, assumptions, and reasoning more transparent.
Following the presentation, the discussion among INDRC SAB members turned to the broader implications of AI for longitudinal Alzheimer’s disease research. Rather than focusing on a single best model, participants emphasised the need for approaches capable of working across complex, multi-modal data collected over time. In this context, they addressed ongoing challenges around data quality, availability, and scale, as well as the difficulty of bringing fragmented data into coherent forms that can be analysed effectively.
Participants stressed that AI systems, while increasingly powerful, must be used with strong validation and human oversight. Broader questions around governance, transparency, trust, and data sovereignty were seen as central to how such systems can be developed and used responsibly. Against this backdrop, agent-based approaches were identified as one promising way to support more transparent, traceable, and reliable workflows, offering a potential pathway for engaging more effectively with complex scientific knowledge.
Taken together, these themes point to a strong potential role for CLARA and the INDRC in shaping emerging standards and frameworks for the responsible use of AI in biomedical research. Insights from the session will contribute to a forthcoming white paper titled “A Scientific Operating System for Multimodal Longitudinal Disease Modeling,” which is currently under preparation and will be published on the CLARA website at a later stage.