Claude finds theoretical flaws in two encryption algorithms
Anthropic has demonstrated that Claude Mythos can identify theoretical mathematical weaknesses in cryptographic algorithms. The model identified flaws in the post-quantum candidate HAWK and a reduced variant of AES.
What happened?
Anthropic has announced that its model, Claude Mythos Preview, has identified theoretical mathematical weaknesses in two cryptographic algorithms. The first discovery halves the effective key strength of HAWK, a post-quantum encryption candidate currently being evaluated by the US-based NIST. The second improves a known attack against a reduced seven-round variant of the symmetric encryption standard AES by a factor of 200 to 800.
Key facts
| Modell | Claude Mythos Preview |
|---|---|
| Målalgoritmer | HAWK och 7-rundors AES |
| Kostnad per upptäckt | Cirka 100 000 USD |
| Prestandaökning AES-attack | 200–800x snabbare |
Why it matters
The findings are notable because they pertain to flaws in the underlying mathematical algorithms, rather than errors in software implementation. HAWK had previously undergone years of manual scrutiny by experts without the weakness being identified. The work was performed largely autonomously and is estimated to have cost approximately $100,000 per discovery.
Who is affected?
This news is primarily relevant to cryptographers, security researchers, and developers of encryption standards. Since the findings are theoretical and do not impact the full version of AES or any active production systems, no end-users or businesses are currently affected.
Impact on the EU
The findings have direct relevance to European authorities and companies preparing for the transition to quantum-resistant cryptographic standards, a central pillar of EU cybersecurity strategies. Future requirements for post-quantum encryption in the EU may be influenced by the ability of AI systems to audit theoretical mathematical models before they are approved for use.
What else you should know
It is important to emphasise that the research results are theoretical and do not imply that current security systems have been compromised. The attack against AES applies only to a simplified seven-round version and does not affect the full algorithm used in practice. However, the research highlights an emerging role for AI in auditing advanced mathematics and security architecture.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Påverkar detta skarpa säkerhetssystem?
The link opens in a new window and leads to the publisher's own site.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
The reader's room
Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.
Sign in to submit a comment or question.
Read the article through your role
- Decide whether this affects strategy over 6–12 months or is just noise.
- Discuss with leadership: do we own the right question or does ownership need to move?
- Ask: what risk are we taking by NOT acting on this this quarter?
Generated angle — not editorial analysis of "Claude finds theoretical flaws in two encryption algorithms"