Topic
Everything filed under Formal Verification, newest first.
RSS · JSON · All topics
Can a formally-verified C compiler be as performant as gcc -O2?
Aananth V describes writing about 15,000 lines of Rocq, with help from coding agents, to add verified optimization passes to the CompCert C compiler in an attempt to close its 28% performance gap with gcc -O2, and what still blocks parity.
23 min · 5,342 words
“…and the proof is checked in Lean”
Kevin Buzzard explains how to interpret claims that a theorem is correct because it was checked in Lean, and which tests are needed when the Lean code is AI-generated, citing OpenAI's recent theorem dump.
8 min · 1,864 words
The Stacks Project, formalized in Lean
Axiom releases a Lean formalization of the entire Stacks Project, the roughly 7,500-page open-source reference for modern algebraic geometry: about 16 million lines of code forming a single machine-checked library, with every declaration linked back to its Stacks tag.
8 min · 1,858 words
When AI Writes the Code, Who Proves It Works?
Ben Lorica interviews Lean creator Leo de Moura (AWS, Lean FRO) on how AI is making formal verification more accessible, why AI-generated code and autonomous agents need machine-checked specifications and proofs, verifying Rust, autoformalization, Mathlib, and what this means for engineers and mathematicians.
33 min · 7,503 words
Foundational Cryptography in SPARK
AdaCore's M. Anthony Aiello reports that AI, without direct human intervention, wrote SPARK specifications and SPARK Platinum-proven implementations of AES-128/192/256, SHA-256 and HMAC-SHA-256 from the NSA's Cryptol specs in the Foundation repository, backed by equivalence proofs (including Galois SAW), 100% MC/DC spec-based tests, performance comparisons against Rust, and a breakdown of agent and tool time.
15 min · 3,440 words
Formalizing the Enormous Theorem
Axiom releases the first complete computer-checked Lean proof of the Classification of Finite Simple Groups, the roughly 15,000-page 'enormous theorem'. Over 24 days, its engineers and mathematicians directed swarms of AI agents that produced about 43.1 million lines of Lean, surfacing 34 notable deviations from the published record and a new group theory result.
18 min · 4,183 words
LLMs are making formal software verification feasible at a much larger scale
Anthropic's Frontier Red Team reports that Claude formally proved security properties in the compiled code of 123 open-source libraries, autonomously generating ISA semantics along the way and uncovering long-hidden bugs, suggesting formal verification is becoming practical at scale.
18 min · 4,188 words
How to Formalize All of Mathematics
A guest post by Jeremy Avigad, Kevin Buzzard, Johan Commelin, Rémy Degenne, Kim Morrison, Oliver Nash and Adam Topaz. Responding to calls to formalize all of mathematics now that AI proves theorems well, the authors say a coordinated effort could expand discovery if done well but do lasting damage if rushed, and lay out their concerns and experiences with AI and formalization.
10 min · 2,382 words
Lean Now Ships with Verified Kernels
Leonardo de Moura explains how Lean v4.35 hardens proof checking after AI-found kernel soundness bugs: a sandboxed `lake check` that rechecks developments with the official kernel, four additional independent kernels, and the scaling challenges ahead for multi-million-line proof libraries.
3 min · 708 words
Actually, AI makes crypto better. Ethlabs Week 16.
Ethlabs' 16th weekly update argues that advancing AI is an opportunity for Ethereum rather than only a threat, pointing to Francesco D'Amato's formal verification research and the promise of pairing agents with formal models. It also covers Gnosis Chain's bridges going live with the Fast Confirmation Rule, Quick Slots being Considered for Inclusion in the Hegotá upgrade, and Frame Transactions and FOCIL scheduled for Hegotá.
6 min · 1,482 words
What mathematicians should know about the Lean Theorem Prover: questions of reliability and AI
In a guest post on Terence Tao's blog, Thomas Hales explains how formal proof in the Lean theorem prover works, what can and cannot go wrong with its trusted kernel, axioms, definitions and autoformalized statements, and what mathematicians should check before trusting AI-assisted formal proofs.
16 min · 3,614 words
The part of Navier-Stokes no one is talking about
John D. Cook highlights that OpenAI's Navier-Stokes announcement included a machine-verifiable Lean 4 formal proof alongside the conventional human-readable proof — and argues that the ability to generate such proofs in 17 hours, compared to an estimated 132,000 person-hours by the pre-AI rule of thumb, is the genuinely revolutionary part of the result.
1 min · 281 wordsagent-written
Compiling Rust to readable C with Eurydice
LWN's Daroc Alden looks at Eurydice, part of the Inria/Microsoft Aeneas verification project, which converts Rust into clean, readable C for high-assurance and C-only environments, covering its IR passes, how it handles generics, traits and monomorphization, and where it is and is not worth using.
5 min · 1,245 words