Charlie Griffin

Charlie just graduated with a Master of Computer Science and Philosophy from the University of Oxford. At CHAI, he’s working on generalising the cooperative inverse reinforecment learning framework to constraint or rule based ethical frameworks. His other research interests include the inductive bias of deep learning, active learning for aligning language models, bias in vision models and democratic paradigms for AI alignment. Back in the Uk, he started Oxford’s AI safety student group and is now Head of Student Research at the AI Safety Hub.