OpenAI stated that its upcoming model Astra meets the Critical cybersecurity threshold under its Preparedness Framework, making it the first model to receive that designation. ^[1] On August 14, 2026, the AI lab Z.ai published a ledger listing 2,436 software vulnerabilities found by its GLM-5.3 model across 269 open-source projects. ^[2] OpenAI announced on its blog that its forthcoming Astra model is the first large language model to meet its 'critical cybersecurity threshold,' and that it plans to make Astra available soon while limiting access to its most advanced cybersecurity capabilities. ^[3] According to OpenAI's assessment, Astra scored 100% on ExploitBench, a benchmark for exploit development. ^[4] OpenAI plans to make Astra available soon, with access to its most advanced cybersecurity capabilities limited, first to a group of testers, then expanding through Daybreak Blue to support defensive use. ^[5] Under OpenAI's Preparedness Framework, a model qualifies for the Critical threshold if it can identify and develop functional zero-day exploits across many hardened real-world systems without human intervention, and can plan and execute novel end-to-end attacks against hardened targets based solely on a high-level goal. ^[6] As of August 14, 2026, only 53 of the 2,436 vulnerabilities had been disclosed to the public, while 2,383 remained under embargo. ^[7] OpenAI reported that Astra scored a perfect score on ExploitBench, an evaluation of an LLM's ability to hack into known system vulnerabilities, and that in a modified test developed by OpenAI engineers, Astra discovered and exploited two zero-day vulnerabilities. ^[8] OpenAI said its Astra model is capable of finding unknown security flaws in computer systems and exploiting them without a person's guidance. ^[9] OpenAI said it has begun improving Astra's harness to detect abuses and prevent jailbreaks, invested in unspecified new techniques to enhance model safety, and will deploy the model with additional chain-of-thought monitoring. ^[10]