AI Optimization Subcore Office Hours and Discussion Forum

Discussion Overview
A major paradox of clinical AI governance can be stated simply: while it appears prudent to validate the clinical and ethical value of AI models prior to their deployment, many limitations and risks emerge only once a model processes data within a specific context of use. Conversely, although post-deployment oversight may enable model recalibration and retraining, such interventions can alter models to such an extent that the original rationale for regulatory validation and licensing no longer holds.

Addressing this paradox is therefore urgent to reduce fragmentation and achieve responsible AI governance.

About the Speaker
Alessandro Blasimme is a bioethicist at the Swiss Federal Institute of Technology (ETH Zurich), where he specializes in ethical and policy issues in biomedical innovation and biotechnology. His work spans translational medicine, precision medicine, digital health, and ageing, with a particular emphasis on the ethical implications of artificial intelligence (AI) in clinical development and health care. His research often blends normative ethical analysis with practical policy guidance and critically examining how existing ethical and regulatory structures can adapt to the challenges posed by AI in health care.

Related links
AI Optimization Subcore Webpage