CyberSIXT Search
Original
Image: SecurityWeek

SecurityWeekEduard Kovacs

Russian Hackers Used Claude to Evade Detection in Spy Campaign

Part of the UNC2452 story · 4 articles · 4 outlets · since 1 Sep 2025

Anthropic has disclosed details of a cyber-espionage operation aligned with the Russian state-nexus group Midnight Blizzard. The attackers reportedly used Claude to monitor how effectively their malware evaded detection, automatically regenerating and redeploying it when security tools flagged it, until it went undetected.

The activity, identified by Anthropic and shut down between December 2025 and August 2026, targeted more than 20 organisations across Ukrainian and European government ministries, defence and intelligence bodies, embassies and think tanks, with additional activity in the Middle East and Asia. Among the seized material, the group exfiltrated mailboxes from two drone-component manufacturers and stole a complete proprietary software development kit for a drone-vision system.

They also compromised at least three hospitality vendors operating hotel guest Wi‑Fi, using stolen admin credentials to redirect guest traffic via DNS hijacking, a method Microsoft has associated with Midnight Blizzard as CaptiveCrunch.

Anthropic’s report also highlights a broader trend: threat actors are increasingly aiming at AI credentials and infrastructure, not just using AI as a tool. In one case, GTG-50021 ran a fraudulent Claude reseller service that proxied paying customers to another model while harvesting Anthropic account credentials for resale.

In a separate instance, GTG-50020, a Russian-speaking group with a history in hotel-booking and fintech breaches, allegedly used prompt injection against an AI vendor’s evaluation sandbox to obtain production API keys for multiple providers. Although the attackers pursued access to a pre-release Claude model, Anthropic says none of the attempts succeeded.

The firm notes that stolen AI credentials can be resold or used to obtain free compute, and urges organisations to treat AI API keys with the same scrutiny as production credentials. The material forms part of a broader threat intelligence report published by Anthropic.

How the story developed

All 4 articles

4 articles from 4 outlets over 376 days. This was the 4th report, 375 days after the first.

4 articles from 4 outlets since 1 Sep 2025; the latest 45 days ago.

UNC2452: coverage over time4 articles from 4 outlets, 1 of them before 25 Oct 2025. First report 1 Sep 2025 by BleepingComputer. Busiest week: 1 article, week of 7 Sep. 0 in the last 7 days. UNC2452: coverage over time4 articles from 4 outlets, 1 of them before 25 Oct 2025. First report 1 Sep 2025 by BleepingComputer. Busiest week: 1 article, week of 7 Sep. 0 in the last 7 days.
  • Article
Show as table
UNC2452: articles
Published (UTC)OutletHeadline
Fri 11 Sep · 08:59 UTCSecurityWeekRussian Hackers Used Claude to Evade Detection in Spy Campaign
Mon 3 Aug · 14:51 UTCInfosecurityAPT29 hijacks hotel Wi‑Fi to spy on travelers via fake updates
Fri 31 Jul · 21:50 UTCMicrosoftMidnight Blizzard Offshoot Hits Travellers With CaptiveCrunch
Mon 1 Sep 2025 · 19:15 UTCBleepingComputerAmazon disrupts Russian APT29 hackers targeting Microsoft 365

Who reported 4 outlets

  • BleepingComputer1
  • Microsoft1
  • Infosecurity1
  • SecurityWeek1
See the full timeline

Sources

On this device

Following

Intel

VendorsThreat actorsResearch reports

Settings

CyberSIXT summarises security news every 30 minutes.

↑↓ move↵ open⌘C copy CVE IDEsc close

Keyboard shortcuts

Everything works with a mouse or touch. Shortcuts are a faster path.

Search ⌘K
Search /
Next item J
Previous item K
Go to Today GT
Go to News GN
Go to CVEs GC
Go to Briefing GB
Reduce motion on this device