AI is being applied to research evaluation faster than the evidence about it is accumulating, and the claims made for it are running ahead of what can be demonstrated. How can we make meaningful use of AI in research evaluation without losing the judgement, context and trust on which good evaluation depends?
In this conversation we no longer question whether AI will be used in research evaluation, but how it can be used in ways that strengthen rather than weaken good judgement.
Our discussion questions include:1. What should we actually use AI for in research evaluation?2. What should we be cautious about asking AI to do?3. What happens to the role of the human evaluator?4. What does trustworthy AI-assisted evaluation look like?
Programme
09:15–09:20 • Welcome & Opening Remarks
Laura Tucker, Hivve
Setting the scene for the conversation.
09:15–09:20 • In Conversation with Wilfred Mijnhardt: AI and the Future of Research Evaluation
Wilfred Mijnhardt and Carlos Galan-Diaz
AI is being applied to research evaluation faster than the evidence about it is accumulating, and the claims made for it are running ahead of what can be demonstrated. How can we make meaningful use of AI in research evaluation without losing the judgement, context and trust on which good evaluation depends?
In this conversation we no longer question whether AI will be used in research evaluation, but how it can be used in ways that strengthen rather than weaken good judgement.
09:15–09:20 • Q&A
Carlos Galan-Diaz
An opportunity to explore the issues raised during the conversation.
09:55–09:59 • Concluding Reflections & Close
Drawing together key insights and considerations from the conversation.
