Anthropic is developing an open-source system called FlashProve designed to verify the authenticity of AI model outputs. The system will allow users to check whether an AI response genuinely comes from the model it claims to be, using cryptographic proofs to ensure the output originates from a specific model. FlashProve aims to prevent AI companies from secretly swapping models to cut costs or exaggerate performance, and can be used to verify third-party claims regarding the use of advanced models for specific tasks. The system requires AI labs to publish model "fingerprints," which do not leak model weights, and the verification process can be executed lightweightly on local devices. Currently, research is ongoing, and neither Anthropic nor other labs have plans to deploy the system yet.