Skip to content
Research · Jul 13, 2026

Microsoft Research open-sources verified Rust cryptography for SymCrypt using Lean and Aeneas

A new methodology combines Rust, the Lean proof framework, and the Aeneas toolchain to formally verify cryptographic algorithms in SymCrypt, with initial proofs for SHA-3 and ML-KEM shipping in Windows Insider builds.

Trust79
HypeLow hype

1 source · cross-referenced

ShareXLinkedInEmail
TL;DR
  • Microsoft Research describes a formal verification workflow that uses Rust, Lean, and Aeneas to prove correctness of cryptographic algorithms in SymCrypt.

Microsoft Research reports a new method for formally verifying cryptographic code written in Rust within SymCrypt, Microsoft’s cross-product cryptographic library used in Windows and Azure. The approach uses the Lean proof framework and the Aeneas toolchain to produce machine-checked proofs that the Rust implementations correctly implement standard algorithms.

The team states that initial verified algorithms include SHA-3 and ML-KEM, and that these verified components are already included in Windows Insider builds. They note that the same Rust, Lean, and Aeneas-based workflow is being extended to additional algorithms such as AES-GCM, FrodoKEM, and ML-DSA for future Windows and Linux releases.

The methodology emphasizes keeping formal specifications close to public standards, enabling line-by-line comparison between the standard text and the Lean model. The team also reports using executable specifications to run against official test vectors to catch transcription errors or misunderstandings of the standard.

To scale verification effort, the team introduces agents that assist in writing proofs and intermediate properties, while compilation, code extraction, and proof verification remain deterministic processes. The open-sourced SymCrypt branch includes both the Rust implementations and the associated formal specifications and proofs.

Sources
  1. 01Microsoft ResearchVerifying Rust cryptography in SymCrypt, from standards to code
Also on Research

Stories may contain errors. Dispatch is assembled with AI assistance and curated by human editors; despite the trust-score filter, mistakes happen. We correct publicly — every article links to its revision history. Nothing here is financial, legal, or medical advice. Verify before relying on any claim.

© 2026 Dispatch. No ads. No sponsorships. No paid placement. Reader-supported via Ko-fi.

Built by a person who cares about honest AI news.