Research
My research on AI resilience, funding in AI safety, and institutions in AI.
Are AI Proofs Messier Than Human Proofs?
On Tuesday, OpenAI released 719 math manuscripts written by an internal model. Some of them solve problems that had been open for decades. Yesterday morning, Tanya and I talked about what this means for the elegance o...
Do Corporate Incentives Teach AI to Protect Itself?
The tools that make an AI safe, like kill switches, also make it stoppable. A model therefore has an incentive to help limit rival AIs but not itself.
