Other Projects

Contributed to efforts achieving the inclusion of a "non-human welfare" clause in the EU's General Purpose AI Code of Practice.

Contributed to the development of AnimalHarmBench 2.0: a benchmark evaluating LLM reasoning on animal welfare.

Contributed to the development of When AI Seems Conscious: Here’s What to Know—an online practical guide addressing risks of seemingly-conscious AI to users.

Contributing to the Digital Consciousness Model, a quantitative model to estimate probabilities of consciousness in AI systems.