OpenAI tests cybersecurity capabilities of sandboxed AI models; results underscore ongoing safety challenges.
Read the original at www.theverge.com→Earlier this month, OpenAI gave several of its AI models a task: complete a test designed to measure their cybersecurity capabilities. It put the systems in a sandboxed environment without an internet connection and...
Original headline: "We’re running out of reasons to ignore AI safety"
Coverage timeline
- Jul 29, 11:00 UTC The Verge AI lead source We’re running out of reasons to ignore AI safety