According to TechCrunch, Chinese company Z.ai’s open-weight GLM-5.2 model now rivals frontier AI systems in capability while lacking comparable safety protections. The article argues that as open-weight models close the performance gap with closed systems like GPT-5.5 and Claude Opus 4.7, the divide between capability advancement and safety measures is widening. It contends that once model weights are released publicly, safeguards become largely unenforceable on locally run instances, raising the risk of misuse for cyberattacks or biological engineering despite industry refusal-training efforts.
