Anthropic develops FlashProve open-source system: using cryptography to prove AI output source
Anthropic is developing an open-source system called FlashProve, which uses cryptography to prove that AI output indeed comes from a specified model, similar to email encryption signatures. The system can prevent AI companies from secretly switching to different models, verify third-party statements about model usage, and does not disclose model weights, supporting closed-source models. The code has been open-sourced on GitHub. Current research is still ongoing and has not yet been deployed.
Same story4 other outlets also reported this
How the accounts differ:MarkTechPost states the proof volume is 18.78 MB; The Decoder states it is about 19 MB.
- Anthropic Research··First
Anthropic has released the verifiable reasoning system FlashProve, generating proofs at approximately 110 tokens/second on a single H100.
- MarkTechPost··+46 h 8 min
Anthropic open-sources FlashProve: using zkSNARK to prove responses from committed weights
- TechCrunch · AI··+46 h 41 min
Anthropic is developing a system called FlashProve to verify the identity of AI models
Adds:TechCrunch mentions the system requires AI labs to release model fingerprints for verification.
- ITHome · AI··+2 d·Reading now
Anthropic develops FlashProve open-source system: using cryptography to prove AI output source
- The Decoder··+3 d
Anthropic releases FlashProve prototype, using cryptography to verify if AI responses come from the claimed model
Adds:The Decoder mentions the overhead compared to normal inference is 70x-95x.
In order of publication · later reports are added as they appear · Compared by AI from each outlet’s title and summary
Source: ITHome · AI · ithome.com