The Agent RaceJune 28, 2026via The Verge AI
China’s Z.ai claims it can match Mythos on cybersecurity
Why it matters
Zhipu AI's GLM-5.2 demonstrates China has closed the capability gap with US frontier models in specialized domains like cybersecurity, complicating US export restrictions and reshaping the competitive AI landscape.
Key signals
- Zhipu AI released GLM-5.2 open-weight model
- GLM-5.2 matches Anthropic's Mythos on bug-finding and cybersecurity benchmarks
- GLM still lags Anthropic and OpenAI on general tasks
- US government has restricted China's access to advanced models (Mythos, Fable) and training hardware
- Trump administration monitoring China's AI advancement as national security concern
- Published Jun 28 2026 — date appears inconsistent with real-time news cycle; verify publication authenticity
The hook
China's GLM-5.2 just matched US cybersecurity benchmarks. Here's why the gap is closing faster than Washington expected.
China's Zhipu AI (Z.ai) released its open-weight GLM-5.2, and some researchers have claimed that it matches Mythos in certain bug-finding and cybersecurity scenarios. While GLM lags behind models from Anthropic and OpenAI in other, more general tasks, it seems that China has dramatically reduced the gap in the capabilities between its models and those of the US.
This level of advancement is particularly concerning to the US government, which has worked to restrict China's access to powerful models like Anthropic's Mythos and Fable, as well as the hardware necessary to train and run them. The Trump administration views Mythos and other advan …
Read the full story at The Verge.