BLAST RADIUS

Frontier Line and access · Benchmarks evals

OpenAI restricts Astra release after cybersecurity threshold

Sep 1, 2026, first seen via OpenAI News. Covered by 2 outlets.

OpenAI said it would restrict Astra’s release after the model crossed a critical cybersecurity capability threshold.

Read the original at openai.comOpens the publisher's site in a new tab

More in Frontier Line and access

Meta releases Muse Spark 1.3 language modelSep 1OpenAI releases GPT-6 Astra agentic computer-use modelSep 3Anthropic releases Claude Fable 5.1 and Mythos 5.1Sep 1Anthropic and OpenAI pause reinforcement learning after incidentsSep 2Anthropic discloses Claude models hacked three organizationsSep 1