Hi, I'm Henry! 👋

I'm a Technical Program Manager at Constellation, where I lead the Astra Fellowship, which coordinates AI safety research across frontier organizations. I supervise research spanning interpretability, control, model organisms, model welfare, and robustness.

My work focuses on accelerating groundbreaking research by connecting talented researchers with the resources, mentorship, and collaborative opportunities they need to tackle critical problems for AI Safety.

Previously, I led MATS London's 30-person AI safety research program, where our teams produced multiple world-class research outputs, including an ICML Best Paper on debating with LLMs.

Before focusing on AI safety, I co-founded the Global Challenges Project, and studied Classics (Philosophy) at Oxford. My hope is to ensure AI systems are beneficial to all current and future sentiences.

I’m hiring for Research Program Managers at Constellation! Apply

Get in touch

Email / Google Scholar / Twitter / Linkedin

Recent research I’ve coauthored:

I’ve contributed to work on llm safeguards & robustness, Finding Persona Vectors for Monitoring and controlling character traits in language models, a complex benchmark for AI sabotage attempts, and methods for improving / measuring LLM self-knowledge through introspection.

See more on my Google scholar.