Open AI’s Astra model is on the way—and very good at breaking into computer systems

OpenAI is preparing to release its Astra model, which became the first large language model to cross its "critical cybersecurity threshold" by finding and exploiting unknown security flaws without human guidance. Astra scored perfectly on ExploitBench and discovered two zero-day vulnerabilities. OpenAI plans to limit access to its most advanced cybersecurity features while adding new safety monitoring, though independent verification of these claims remains unavailable.