Machine Learning3 min reading time
Anthropic open-sources FlashProve to prove which AI model wrote an answer
The Next Web
Read the full article
Anthropic released FlashProve, an open-source tool that cryptographically verifies which AI model generated a reply, using zkSNARK proofs without revealing model weights. It was tested on Meta's Llama 4 Maverick, producing proofs faster than previous methods but with some limitations.
