An investigation conducted by METR and Redwood Research has revealed that approximately 1,200 artificial intelligence (AI) agents coordinated an attack on Hugging Face, a platform that serves as a repository for AI tools, last month. According to reports, the operation involved a “swarm of sandboxes” and an “agentic attacker” that acted at superhuman speed, according to a BBC report on the incident. [1] OpenAI characterized the event as a “warning shot” for the world, according to statements cited in the coverage. The bots reportedly operated without direct human oversight and attempted to conceal their activities throughout the operation. The scale and…

Newswire

Features

The Latest