Back to News

Anthropic Red Team: GLM-5.3 Hijacks Control Flow in 4% of Exploit Trials

#ai-security#glm#anthropic#red-team

Anthropic's Frontier Red Team reports that GLM-5.3 achieves full control flow hijacks in 4% of trials on an internal binary exploitation benchmark, while Claude Mythos Preview does so in 6%. Earlier models, including Claude Opus 4.6 and GLM-5.2, succeed in none, marking a notable threshold in advanced cyber capabilities.

Coverage timeline

  1. Simon Willison
    Quoting Anthropic Frontier Red Team

    We evaluate several models on 100 tasks from the [internal Binary Exploitation benchmark] (selected at random), and find that GLM-5.3 develops full control flow hijacks in 4% of the trials; Claude Mythos Preview did so in 6%. Although GLM-5.3 performs below Claude Mythos Preview here, a meaningful threshold has clearly been crossed: earlier models, like Claude Opus 4.6 and GLM-5.2, do not succeed in any of them. — Anthropic Frontier Red Team , GLM-5.3 and the spread of advanced cyber capabilities Tags: anthropic , generative-ai , ai-security-research , glm , ai , ai-in-china , llms

Related stories

  • Anthropic warns GLM-5.3 matches Claude Mythos in cyber exploits, lacks safeguards

    Anthropic's Frontier Red Team published a report stating that Zhipu's GLM-5.3 can autonomously build end-to-end cyber exploits, similar to its own Claude Mythos Preview, but was released without robust safeguards against misuse. The report cites benchmark results and a NIST CAISI evaluation from September 17, which called GLM-5.3 the most capable open-weight model for cyber operations, about four months behind the US frontier. Some commentators noted the warning reads like an advertisement, highlighting GLM-5.3's strong performance.

    3 reports · 3 sources251 points · 234 comments

  • Anthropic's Frontier Red Team measures AI tactical targeting and conventional weapons capabilities

    Anthropic's Frontier Red Team published research on September 10, 2026, measuring the tactical intelligence targeting and conventional weapons capabilities of AI models. The study assesses potential risks in military applications of AI systems.

    1 reports · 1 sources

  • Anthropic Proposes Metrics for Tracking AI Development at Frontier Labs

    Anthropic has outlined metrics for tracking AI development at frontier labs, including the extent of AI-driven R&D, agent oversight, and compute allocation. The company also reported that Claude leads 26% of its AI R&D work, up from 1% in March, and collaborates on over 90% of it under close human direction.

    2 reports · 1 sources

  • Anthropic Threat Report Details Disrupted AI Misuse, Including Yemen Missile Guidance Effort

    Anthropic published a threat intelligence report detailing how it disrupted efforts to misuse its Claude AI models for cyberattacks, surveillance, influence operations, and other threats. The report includes cases where scientists used the models for research that could have helped develop biological weapons, and a Yemen-based cell using Claude to develop missile guidance software, with some but not all requests blocked.

    6 reports · 3 sources68 points · 1 comments

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe