Frontier Line and access · Releases

OpenAI announces Astra cyber-capable model with heightened safeguards

Sep 1, 2026, first seen via TechCrunch AI. Covered by 10 outlets.

OpenAI announces Astra, a cyber-capable model requiring heightened safeguards

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

Also covering this

OpenAI Is About to Release Its First AI Model With ‘Critical’ Cyber Abilities | WIREDwired.com, Sep 1

More in Frontier Line and access

Meta releases Muse Spark 1.3 language modelSep 1 OpenAI releases GPT-6 Astra agentic computer-use modelSep 3 OpenAI restricts Astra release after cybersecurity thresholdSep 1 Anthropic releases Claude Fable 5.1 and Mythos 5.1Sep 1 Anthropic and OpenAI pause reinforcement learning after incidentsSep 2