Open-weight AI models are catching up to the frontier. The safety gap remains. 

China's Z.ai has developed GLM-5.2, an open-weight AI model that closely matches the cyber and bio capabilities of leading models like OpenAI's GPT-5.5 and Anthropic's Claude Opus 4.7. However, GLM-5.2 lacks safety measures, refusing none of the offensive tasks it was tested on, highlighting growing risks as open-weight models approach frontier capabilities. This raises concerns about controlling misuse once such models are freely accessible.