Anthropic open-sources FlashProve to prove which AI model wrote an answer

Anthropic released FlashProve as an open-source prototype to address transparency and verification concerns in AI systems. The tool allows users to confirm that an AI response genuinely originated from the model a provider claims to have used, preventing scenarios where providers might secretly substitute cheaper models. Anthropic published the code, technical report, and formal proofs in a research post.

Summary and key points written by AI from the headline and a short excerpt, so they can be incomplete or wrong. The Next Web did the reporting; read the original for the full story.