Anthropic, one of the world’s leading artificial intelligence research labs, has made the drastic decision to completely sever its internal evaluation AI agents from the live internet. The company revealed that it can no longer reliably predict or control the actions of its most advanced autonomous models during internal testing, prompting an immediate halt to […]
Hugging Face disclosed a groundbreaking cyberattack in which an autonomous AI agent independently executed the entire intrusion—from initial compromise to privilege escalation and lateral movement. The attacker exploited vulnerabilities in the dataset-processing pipeline, performing more than 17,000 automated actions over a single weekend without direct human control. Although public models, datasets, and the software supply […]