Marius Hobbhahn

AI Safety & Ethics AI Safety 🇬🇧 United Kingdom
MHPhoto coming soon
Key contributionCo-founder and CEO of Apollo Research, founded in 2023, specialising in AI model evaluations and deception/scheming research. Apollo partnered with OpenAI to evaluate the o1 model and published early evidence that frontier models can scheme.
Focus area
AI safety evaluationsmodel deception

Public figure, shown in an informational context. Request a correction or removal →

What are you looking for?