# Mythos Finds HAWK and 7-Round AES Weaknesses—Research Milestone, Not Bank Break

Times of AI Desk · 2026-07-28 · Research

[https://timesof.ai/2026/07/anthropic-mythos-cryptographic-weaknesses](https://timesof.ai/2026/07/anthropic-mythos-cryptographic-weaknesses)

> Claude Mythos Preview improved a key-recovery attack on the HAWK post-quantum signature candidate and a 200–800× faster attack on seven-round AES-128—mostly autonomously at ~$100k API cost each. Neither breaks production systems today; verification cost is the scarce resource.

This is a **capability milestone for AI-assisted cryptanalysis**, not a “your bank is broken” headline. The durable points: frontier models can contribute **publishable** crypto research under human verification bottlenecks; NIST-style public contests are where such attacks belong **before** deployment; verification cost is becoming the scarce resource.

**Anthropic** published **“Discovering cryptographic weaknesses with Claude,”** reporting that **Claude Mythos Preview** found **improved mathematical attacks** on two targets: **HAWK** (a NIST post-quantum signature candidate) and a **seven-round reduced** variant of **AES-128**. Results were developed **mostly autonomously** (~**$100,000** API cost each; HAWK ~**60 hours**), disclosed to algorithm authors, NIST channels, and U.S. government/industry partners. Anthropic stresses **neither result breaks production systems** today.

## The two primary results

| Target | What Mythos improved | Production impact |
|--------|----------------------|-------------------|
| **HAWK-256** (NIST PQC additional-signature candidate) | End-to-end **key-recovery** via a previously unexploited lattice automorphism; effective key strength roughly **halved** (e.g. small parameter cost cited as **2^64 → 2^38**) | **None deployed** — HAWK is a **candidate**, not a fielded standard; larger parameters remain impractical to attack under this exponential improvement |
| **7-round AES-128** (research-only reduced cipher; full AES-128 has **10** rounds) | New **meet-in-the-middle** fingerprint (“**Möbius Bridge**”) claimed **200–800×** faster than prior best under chosen-plaintext assumptions | **Does not break full AES**; prior-style attacks already assume huge chosen-plaintext budgets |

Anthropic also notes preliminary work on reduced-round **LEA**, **Serpent**, and limited gains on **Salsa20**, **Poseidon**, and **SHA-1**, with fuller write-ups planned after verification.

## How the research was run

- **HAWK**: semi-autonomous multi-agent harness (Claude Code–like) with occasional human project management; human operator not a lattice specialist; end-to-end verification pipeline included.
- **AES**: scaffold for hypothesis → experiment → attack design; model initially refused (“AES is too studied”); a handful of nontechnical prompts over days, then roughly **one billion** output tokens refining the attack; humans spent **hundreds of hours** validating.
- **Artifacts**: technical PDFs (HAWK + AES), AES chain-of-thought extract, demo code, and **CryptanalysisBench** (with ETH Zurich, Tel Aviv University, University of Haifa / TU Berlin academic partners per Anthropic posts).

## Limits

- Speedup and key-strength figures are Anthropic’s research claims pending community cryptanalysis review.
- Full AES-128 (10 rounds) is explicitly not broken.
- HAWK is a candidate; parameter-set implications for NIST process are ahead.

## Sources

- [Anthropic: Discovering cryptographic weaknesses with Claude (July 28, 2026)](https://www.anthropic.com/research/discovering-cryptographic-weaknesses)
- [Anthropic HAWK key-recovery paper (PDF)](https://www.anthropic.com/document/hawk_key_recovery.pdf)
- [Anthropic AES Möbius Bridge paper (PDF)](https://www.anthropic.com/document/aes_mobius_bridge.pdf)
- [CryptanalysisBench (arXiv:2607.18538)](https://arxiv.org/abs/2607.18538)
- [The Hacker News / NYT same-day coverage of Mythos cryptanalysis](https://thehackernews.com/2026/07/claude-ai-just-cracked-post-quantum.html)