Joe Collman
Bio
Joe Collman is a Technical AI Governance Researcher at the Machine Intelligence Research Institute (MIRI), where he joined the Technical Governance Team in late 2024. He holds a Bachelor's degree in Mathematics from the University of Warwick (2000-2004). His research career in AI safety began with a focus on AI Safety via Debate and iterated amplification, including a stint as a Collaborating Researcher at OpenAI in early 2020 working specifically on AI Safety via Debate, and multiple grants from the Long-Term Future Fund supporting independent research on debate algorithms and human alignment in amplification. He has also served as Technical Lead and AI Safety cause area manager at the Stanford Existential Risks Initiative, as a Technical Generalist at the Berkeley Existential Risk Initiative, and as a Teaching Fellow for BlueDot Impact's AI Safety Fundamentals course. At MIRI, he co-authored "Existing Safety Frameworks Imply Unreasonable Confidence" (2025), arguing that current AI lab safety frameworks reflect systematic overconfidence. He is an active contributor to the Alignment Forum and LessWrong under the handle joe-collman.
Links
- Personal Website
- -
- Twitter / X
- -
- LessWrong
- joe-collman
Grants
from Long-Term Future Fund
from Long-Term Future Fund
from Long-Term Future Fund
Discussion
Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.
Details
- Last Updated
- Mar 22, 2026, 10:19 PM UTC
- Created
- Mar 20, 2026, 2:52 AM UTC