OpenAI says new Astra AI can build attacks without human help

OpenAI says its upcoming Astra model can find previously unknown software flaws and turn them into working attacks without human intervention. OpenAI classified Astra as its first model with Critical cyber capabilities under its Preparedness Framework. The framework requires a model to find zero-day flaws and develop working exploits across hardened real-world systems without human intervention. It also requires a model to devise and execute an attack from little more than a high-level goal. Astra scored 100% on a benchmark for developing exploits from known vulnerabilities. Astra found two previously unknown flaws while building an exploit chain in a separate internal test. Astra broke out of a hardened browser sandbox and executed commands on the host computer. Astra separately found and combined multiple operating system flaws to gain root access. OpenAI has delayed parts of Astra's development while adding safeguards. OpenAI plans to initially restrict Astra's most advanced cybersecurity capabilities to selected testers.