OpenAI previews Astra, hitting Critical threshold in cybersecurity capability

arthurcolle · x · 2026-09-02

OpenAI officially previewed the upcoming Astra release, describing it as a significant advance in cybersecurity capability that reaches the Critical threshold under its Preparedness Framework. The company also previewed how it evaluated the model, how safeguards advanced alongside capabilities, and what it will keep improving.

Related event: OpenAI to Release Astra, Its First Model Hitting Critical Cybersecurity Threshold(9 posts)→

Original post →

More from Models

Models channel →