Technologyai
OpenAI’s Math Solutions Fall Short of Academic Standards
OpenAI recently released hundreds of solutions to complex mathematical problems, claiming to have consulted elite mathematicians. However, researchers noted that the lab missed several key advisory guidelines. Critiques center on the lack of human understanding and transparency in the AI-generated proofs. Only a fraction of the manuscripts included the model's chain of thought, and many proofs lacked formal verification.