CAIF is delighted to welcome Thore Graepel to our Board of Trustees. Thore was an author on two influential papers that launched cooperative AI as a field, and he has published widely on the topic of multi-agent learning. Thore has recently returned to Google DeepMind, where he leads a team working on post-AGI futures.
Previously, Thore was global lead for Computational Sciences, AI and Machine Learning at Altos Labs, which works to understand cellular rejuvenation programming with the goal of reversing disease to transform medicine. In an earlier role at Google DeepMind, Thore contributed to the development of AlphaGo, the AI system that mastered the ancient game of Go and inspired a new wave of AI applications to solve complex problems. Thore is also well known in the data privacy field through his work on the predictability of private attributes from digital records of human behaviour.
Thore has served as a member of the Board of Directors at the Partnership on AI (PAI), and alongside his other roles he is part-time Chair of Machine Learning at UCL’s Department of Computer Science. As a champion of responsible innovation in AI, Thore sees the establishment of cooperative social skills among artificial agents as fundamental, and this is central to his research on multi-agent learning. Thore brings a rich portfolio of experience to the CAIF Board at a critical time in the emergence of a world of interacting AI agents.
October 30, 2025
