Ed Hughes

Edward Hughes is Chief Scientist at Inherent, a London-based lab building AI that discovers new knowledge. In his former role, he co-led the AI Scientist team at Google DeepMind, wrote the final implementation of the Genie 1 world model, pioneered cultural evolution among LLMs, and contributed to AlphaProof.

Over the past decade, Edward’s teams have spearheaded fast adaptation in reinforcement learning, the paradigm of Cooperative AI, and ad-hoc collaboration between machines and humans. He draws inspiration from diverse sources, including ethics, cultural evolution, social psychology, economics, organisational design, and meta-learning.

Edward is a Visiting Fellow at the London School of Economics, an Advisor to the Cooperative AI Foundation, and a PIBBSS mentor. He was previously a Staff Research Scientist and a Staff Research Engineer at Google DeepMind. He received his PhD in Theoretical Physics from Queen Mary University of London on applications of string theory to particle scattering.

Edward’s publications are available on Google Scholar.