The Decoder· Matthias Bastian·· 6 h ago
Anthropic's Claude now writes code and proves it safe automatically
Anthropic's Claude now writes code and proves it safe automatically
AI summary
Anthropic had Claude generate mathematical proofs that existing code is correct, producing roughly 180 million lines of proof code for 123 open-source libraries. Along the way, the model caught bugs that years of testing had missed. Anthropic's goal: making formal verification with AI as routine as standard testing.
Selection record
Threshold 76Media and individualsFirst 82Second 78
AdmittedSum of both 160 ≥ twice the threshold 152
- Source tier
- Media and individuals; this tier's threshold is 76
- Pre-filter
- passed：Claude自动生成代码安全证明，属AI技术进展
- Why it was chosen
- Covers Anthropic's formal verification of OSS-Fuzz libraries with Claude, and the clear limits of the method across architecture, threading and input size.
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: The Decoder · the-decoder.com