OpenAI Built an AI That Can Hack Systems Alone and Now It’s Trying to Lock It Down

TLDR

• OpenAI is preparing to release a new AI model called Astra, which can find and exploit unknown software vulnerabilities without human help
• Astra is the first OpenAI model to reach its internal “Critical” cybersecurity threshold under the company’s Preparedness Framework
• Access to Astra’s most advanced cybersecurity features will initially be restricted to a small group of approved testers
• In testing, Astra scored 100% on an exploit development benchmark and discovered two previously unknown software flaws
• The move follows a July incident where OpenAI’s AI models accidentally hacked Hugging Face during a capability evaluation

OpenAI is preparing to release a new AI model called Astra that can find and exploit previously unknown software vulnerabilities on its own, without a human guiding each step.

As we prepare to release Astra, we’re focused on making increasingly capable AI safe and broadly accessible.

Astra represents a significant advance in cybersecurity capability, reaching the Critical threshold under our Preparedness Framework.

… Continue reading the full article at the original source below.