https://derekwang2002.github.io/ai-intelligence/en/projects/oss-verify/ Material changes in AI models, agents, engineering and infrastructure, with primary sources and recommendations. en Mon, 02 Nov 2026 16:56:00 GMT https://derekwang2002.github.io/ai-intelligence/en/events/ev-20261022-06/ ev-20261022-06-published Fri, 23 Oct 2026 16:04:55 GMT New TRIAL <p>Anthropic and Theorem released MIT-licensed Rocq proofs and verifier components for selected C library fuzz entry points. Anthropic reports a larger internal run, while the public proof repository currently covers 60 entry points across 25 libraries. The guarantee is bounded to modeled static, single-threaded RISC-V Linux processes and each fuzz target&apos;s stated properties.</p><p><strong>Why read: </strong>Public proofs can be replayed, while internal scale, the public subset and real software stacks remain distinct.</p><p>Recommendation: TRIAL</p><p>Sources: <a href="https://www.anthropic.com/research/formal-verification">Using AI and formal verification to secure open-source software</a> · <a href="https://github.com/theorem-labs/oss-verify-proofs">oss-verify-proofs</a> · <a href="https://github.com/theorem-labs/rocq-linux-process-semantics">Rocq Linux process semantics</a></p>