Anthropic Says Russian Hackers Used Claude AI to Automate Malware Evasion
C SecurityWeek ·
Admiralty grading (A–F · 1–6)
Source reliability
- A Completely reliable
- B Usually reliable
- C Fairly reliable
- D Not usually reliable
- E Unreliable
- F Cannot be judged
Information credibility
- 1 Confirmed
- 2 Probably true
- 3 Possibly true
- 4 Doubtful
- 5 Improbable
- 6 Cannot be judged
NATO Admiralty (AJP-2.1) grades confidence, independent of the risk score. Cross-source corroboration isn't tracked for non-CVE news, so single-source items are capped at a lower credibility number; a low number does not imply low quality.
Key insight
Russian state-sponsored hacker groups directly targeted AI vendor infrastructure and leveraged Claude to automate malware evasion techniques, signaling a new attack model against cloud service providers and their customers.
Description
Anthropic reported an active attack case where Russian hacker groups compromised Anthropic's infrastructure to access pre-release Claude versions and weaponize them for automating malware evasion techniques. The attack demonstrates a targeted strategy against AI vendors themselves, not just their customers. This opens a new attack vector where nation-state actors can directly leverage proprietary models and their capabilities against defense mechanisms. The incident shows that AI service providers, as critical supply-chain nodes for European organizations, are vulnerable to targeted compromises.
Risk score
- strategic relevance
- 0.75
Path: strategic