More OpenAI math ‘proofs’ www.reddit.com/r/…/more_openai_math_proofs/ they claim they got halfway to solving another millennium prize problem, but issues with Lean translation make it sketchy at best