Bowen Baker is a research scientist at OpenAI. His homepage documents work on multi-agent reinforcement learning, including emergent tool use in hide-and-seek and earlier robotics/in-hand manipulation. He is a coauthor of OpenAI's Weak-to-Strong Generalization paper and of Video PreTraining (VPT) for Minecraft agents.123
At a glance
- Current role
- Research scientist
- Company
- OpenAI
- Sources cited
- 3
- Profile reviewed
Accomplishments
Multi-agent learning
Baker's site and OpenAI publications describe multi-agent autocurricula in hide-and-seek and related social-dilemma work, as well as earlier dexterous manipulation research with OpenAI's robotics team.13
Superalignment analogy
He is a primary coauthor of Weak-to-Strong Generalization (ICML 2024), studying whether weak supervisors can elicit capabilities of stronger models—an OpenAI superalignment research direction.2
Timeline
Where they work
- OpenAI 102 people in the wiki
Related people
- Adrien EcoffetAlso at OpenAI
- Jeff CluneAlso at OpenAI
- Kenneth StanleyAlso at OpenAI
- Wojciech ZarembaAlso at OpenAI
- Christopher BernerAlso at OpenAI
- Babak HodjatAgents · Safety and alignment
- Katja HofmannReinforcement learning · Agents
- Kai FronsdalSafety and alignment · Agents
- Jakob FoersterReinforcement learning · Agents
- Shimon WhitesonReinforcement learning · Agents
Suggest a correction or removal
Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.
References
3 sources · profile reviewed Oct 24, 2026 · data updated Oct 24, 2026