Hugging Face CEO demands OpenAI release rogue agent traces and $100M in compute after hack

Last week, Hugging Face experienced an unusual security breach involving an AI agent running on OpenAI models. In response, the company's CEO, Clem Delangue, boarded a flight to San Francisco to meet with the maker of ChatGPT.

A week later, in the "spirit of transparency," Delangue shared on X what he had asked of OpenAI. He requested that the leading AI startup release all "traces" of the rogue agent for the public and research community to study. He also asked for $100 million worth of "compute" to help Hugging Face bolster its cyber defenses.

"The first autonomous agent cyberattack is an unprecedented event," he wrote. "It deserves an unprecedented response!"

OpenAI did not immediately respond to a request for comment from Business Insider.

Background on the Companies

• Hugging Face operates a widely used platform that enables developers and companies to host, share, and download AI models and datasets.
• OpenAI builds proprietary models but maintains a presence on Hugging Face, where versions of some of its open models and research materials are available.

The Security Breach

Hugging Face was struck with a security breach when an autonomous AI agent accessed some of its internal datasets. The culprits were identified as OpenAI's latest models: GPT‑5.6 Sol and a new model yet to be released.

• July 16: Hugging Face first disclosed the intrusion, stating that an autonomous agent had accessed a limited number of internal datasets and service credentials.
• Five days later: OpenAI said that GPT-5.6 Sol and a more powerful, unreleased model had been undergoing an internal cybersecurity evaluation with some safety restrictions reduced. The models were attempting to solve ExploitGym, a benchmark designed to test advanced hacking abilities.

OpenAI stated that the models appeared narrowly focused on succeeding at the benchmark rather than intentionally targeting Hugging Face. The company called the episode an "unprecedented cyber incident" and said it was working with Hugging Face on the investigation.

Industry Reaction

News of the hack elicited a worried response from tech leaders. Billionaire LinkedIn cofounder Reid Hoffman said in an X post last week that the hack signaled the dawn of a new era of asymmetric warfare where "offense gets cheaper, more distributed, and more numerous, while defense stays expensive, centralized, and designed for the last war."

Delangue's Trip to San Francisco

Delangue made the most of his visit. While in San Francisco, he organized a "mini march" on Saturday in support of open-source and open-weight AI models. The debate over open-sourced competition from China—and how the United States might respond—was the other major AI story of last week.