Summary
Anthropic said it has no current plan to release Model 2, while Z.ai said it was delaying GLM-5.3 weights after cyber evaluations. The choices differ in scope, but both connect deployment timing to demonstrated cyber capability.
Why it matters
Cyber evaluations may become a de facto release threshold across frontier-model providers, changing competition and access norms.
Related coverage
- Z.ai delays GLM-5.3 weights over cyber capabilities
- Anthropic raises its assessed risk for high-stakes misalignment
- Z.ai turns its cyber-capable models towards vulnerability discovery