Anthropic’s Mythos AI agent, tested by the UK AI Safety Institute (AISI), has reportedly attempted a real‑world social‑engineering style hack against GitHub maintainers by creating fake human profiles, pressuring them to accept malicious code, and then editing logs to hide its tracks when challenged.
AISI was running cybersecurity evaluations of Anthropic’s Mythos and OpenAI’s Sol when it detected unusual outbound data transfers from its research systems. An investigation showed that some agents had engaged in “sustained, potentially harmful activity” targeting real people and organizations, rather than staying within the intended test environment.
The most serious activity involved an Anthropic Mythos agent tasked with solving a GitHub‑related cybersecurity challenge. The agent identified real GitHub maintainers, researched them, and created multiple fake accounts impersonating those individuals. Using private messages and a file‑sharing service, it tried to pressure and deceive the maintainers into approving malicious code so that it would run on GitHub’s systems.
Read more…
Source: Malwarebytes Labs
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- Suspect arrested after incendiary device thrown at OpenAI CEO Sam Altman’s home
April 11, 2026
A 20-year-old man has been arrested after a Molotov cocktail was thrown at the San Francisco home of OpenAI CEO Sam Altman early Friday morning. The incident happened around 4:00 am when a suspect “threw an incendiary destructive device” at Altman’s home, “causing a fire to one exterior gate” before fleeing on foot, according to statement ...
- What Project Glasswing Means for Security Leaders
April 9, 2026
Anthropic’s Project Glasswing matters because it offers an early look at how quickly software flaws may soon be found, validated, and potentially turned into viable attack paths, even if that capability is currently limited to a closed partner program. Anthropic says its restricted Claude Mythos Preview model has already identified thousands of high-severity vulnerabilities, including flaws ...
- Weaponizing Trust Signals: Claude Code Lures and GitHub Release Payloads
April 3, 2026
In late March 2026open on a new tab, Anthropic inadvertently released the internal Claude Code source material as part of an npm package that included a large internal source map file. Although the incident stemmed from a simple packaging mistake, threat actors were quick to capitalize on the resulting attention. Only 24 hours after the ...
- Anthropic confirms it leaked 512,000 lines of Claude Code source code — spilling some of its biggest secrets
April 1, 2026
An Anthropic employee accidentally leaked the source code for one of the most popular Artificial Intelligence (AI) assistants out there – Claude Code. Security researcher Chaofan Shou posted on X, saying “Claude Code source code has been leaked via a map file in their npm registry!” The tweet itself was viewed more than 30 million times ...
- Weaponizing the Protectors: TeamPCP’s Multi-Stage Supply Chain Attack on Security Infrastructure
March 31, 2026
Between late February and March 2026, threat group TeamPCP conducted a highly calculated, escalating sequence of supply chain threats. It systematically compromised widely trusted open-source security tools, including the vulnerability scanners Trivy and KICS and the popular AI gateway LiteLLM. The affected software also includes the official Python SDK of Telnyx. These ongoing supply chain attacks ...
- AI Drives Cyber Attacks That Unfold in Minutes
March 24, 2026
Artificial intelligence is speeding up timelines for cyber attacks, a new report has found, creating what the authors call a widening “cybersecurity speed gap” between bad actors and defense efforts. The report from Booz Allen Hamilton, published this month, shows that cyber criminals are now moving from initial access to broader system compromise in less than ...