Jan Kirchner
Bio
Updated 03/22/26Jan Hendrik Kirchner is a researcher at Anthropic working on AI alignment and safety, focused on scalable oversight methods and ensuring AI systems behave reliably. He previously worked as a research engineer at OpenAI (2022–2024), where he co-authored the influential "Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision" paper. He holds a PhD in computational neuroscience from the Max Planck Institute for Brain Research in Frankfurt, Germany. His connection to AI safety research began through participation in AI Safety Camp (AISC6, 2022), where he worked on a project analyzing the AI alignment research landscape. He is active in the AI safety community through his Substack newsletter "On Brains, Minds, And Their Possible Uses" (universalprior.substack.com), as well as contributions on LessWrong and the AI Alignment Forum under the handle "jan-2". His work sits at the intersection of computational neuroscience, language models, and alignment research.
Community Signal
Updated 03/22/26No endorsements yet.
Links
Updated 03/22/26- Personal Website
- https://universalprior.substack.com/
- Twitter / X
- LessWrong
- jan-2
- EA Forum
- -