About Serre AI
Serre AI is an independent AI research lab in NYC, working on the formal foundations of AI reasoning.
Mission
We study reasoning in AI systems with the tools of computational complexity and verification, and test what that theory predicts through large-scale empirical evaluation of language models.
The lab is named after Jean-Pierre Serre, the mathematician, for the clarity and rigour of his work. He has no connection with the lab.
Research
Our research is organised in three standing programs:
- Computational complexity of reasoning
- Which reasoning problems are hard for language models, and why?
- Verification
- When can an answer be checked more cheaply than it can be produced?
- Large-scale empirical evaluation
- Does the theory hold when it is tested on real models?
How we work
Rigour is the method. Every claim we make is tied to its evidence. We favour theory that makes testable predictions, and we test those predictions at scale.
We build our own research infrastructure, Serre, to hold the research process to the same standard. About our infrastructure
Leadership
- Oddur Sigurdsson
- Founder
Contact
We welcome researchers interested in collaboration. Write to us at contact@serre.ai.