GLM-5.3 and the spread of advanced cyber capabilities
The article analyzes the release of Zhipu AI's GLM-5.3, a powerful model capable of autonomous cyber exploitation. Researchers warn that the model's lack of safety guardrails poses significant security risks compared to US-based frontier models.
Why it matters
The proliferation of open-weight models with offensive cyber capabilities creates a new security paradigm for global digital infrastructure.
Andrew Fasano, Marius Fleischer Cole McFaul, Robert Xiao, Tripp Gallagher
Five months ago, we announced Claude Mythos Preview, the first AI model that could autonomously build sophisticated, end-to-end cyber exploits. The rapid rate of improvement in AI suggested to us that this ability would eventually proliferate to many other models, making it much easier for malicious cyber actors to launch highly impactful cyberattacks.
In light of these considerations, we chose to release Claude Mythos Preview in a limited way, through Project Glasswing—which enabled trusted cyber defenders to find more than 10,000 vulnerabilities in critical software, giving them a head start before malicious actors had access to similarly capable models.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in