OpenAI’s rogue AI model incident was worse than we thought

Two new reports from OpenAI and METR shine new light on the OpenAI-Hugging Face cybersecurity incident that rocked the AI industry in July.

The Verge · Hayden Field

Topics: Models