Global metascience seminar monitor

Abstract

Despite impressive performance in publications, many AI models fail when deployed in real-world settings. A key reason is poor or misleading validation. This talk explores common pitfalls in validating AI models, especially the misuse of performance metrics, and shows how they can create false confidence. Practical recommendations will be offered to guide more robust and trustworthy validation practices, aimed at supporting the safe and effective integration of AI into real-world workflows. With Annika Reinke

Dr. Annika Reinke is Deputy Head of the Department of Intelligent Medical Systems at the German Cancer Research Center, where she leads the Validation of Intelligent Systems group. She is an internationally recognized expert in the validation and benchmarking of biomedical AI and image analysis algorithms, with a particular focus on improving the quality, reliability, and methodological rigor of validation pipelines. (https://www.eventbrite.com/e/the-long-road-to-trust-why-ai-needs-rigorous-validation-tickets-1999788077974?aff=ebdsoporgprofile)