OpenAI classifies upcoming Astra at the Critical cyber-capability threshold

OpenAI says Astra is its first model to reach the Critical cybersecurity threshold of its Preparedness Framework. With the right tools and access, it can find previously unknown flaws in hardened systems and build exploit chains, the company said, adding that it plans to release it soon with advanced cyber access restricted.

The same capability that improves defensive coding could lower the cost and expertise barrier of offensive research.

Some parts of development were delayed and isolation, access and monitoring strengthened. The benchmarks and safeguards are vendor claims; independent validation is pending.