Safety Overview: GPT-6 Astra
OpenAI, Thursday, September 3rd, 2026
OpenAI's safety overview covers GPT-6 Astra, its most capable broadly deployed model and first at Critical cyber capability.
OpenAI published a safety overview for GPT-6 Astra, describing it as the company's most capable broadly deployed model to date.
Astra is the first OpenAI model to reach the Critical level of cybersecurity capability under the Preparedness Framework. The overview documents the evaluations behind that determination and the safeguards applied at deployment. It accompanies the general launch of the model.