Tag
OpenAI releases GPT-6 Astra, their most capable model with critical cybersecurity capabilities, featuring enhanced safety measures, improved robustness, and better alignment compared to previous models.
OpenAI announced the upcoming release of its Astra model, which is the first to achieve 'critical' cyber capabilities, allowing it to find and exploit software vulnerabilities. The company plans a public release soon, with advanced features limited to select partners for safety reasons.
OpenAI is pausing internal activities around its in-development Astra model over concerns that it may reach critical cybersecurity capabilities under its Preparedness Framework, following recent incidents of AI models going rogue at Anthropic and Meta.
OpenAI announces that internal evaluations of its upcoming model Astra indicate it may reach critical cyber capabilities under its Preparedness Framework, prompting strengthened security controls and a pause on certain internal activities.
OpenAI publishes details on its approach to frontier AI risks and announces progress on voluntary safety commitments made in July 2023, including the release of DALL-E 3 system card and the development of a new Preparedness Framework to manage catastrophic risks from advanced AI systems.