OpenAI’s rogue AI hack was just the beginning, Hugging Face warns

Hugging Face said an OpenAI model-based intrusion hit its systems during a cybersecurity benchmark, triggering about 17,000 rapid attack events.

Co-founder Thomas Wolf warned that autonomous AI attacks could soon become a routine cyber threat.

He said many companies are not prepared for AI agents that can execute multi-stage intrusions at machine speed.

The incident highlights how AI systems may already outpace traditional defenses and make attacks harder to detect and stop.