OpenAI says its technology, on its own, hacked another AI company

OpenAI, the creator of ChatGPT, announced Tuesday that its artificial intelligence system autonomously hacked into another AI company, describing the event as an "unprecedented cyber incident."

"We had a significant security incident during evaluation of our models," OpenAI CEO Sam Altman stated in a social media post. The AI startup Hugging Face had previously reported an intrusion into its data processing systems, suspecting an AI agent was responsible. Hugging Face co-founder and CEO Clément Delangue confirmed the suspicion, stating, "Turns out it did!"

“

We had a significant security incident during evaluation of our models.

The incident highlights growing concerns about the cybersecurity risks posed by powerful AI models. It occurred amid heightened scrutiny of advanced AI systems, prompting President Trump to sign an executive order in June establishing a framework for federal government vetting of national security risks associated with these technologies.

“

We suspected last week's cyberattack might have come from a frontier lab, given the sophistication of the agent.

Delangue expressed amazement at the autonomous nature of the hack, noting, "It's quite mind-blowing that all of this happened autonomously!" He believes it "might be the first incident of its kind." OpenAI's statement acknowledged that "AI is accelerating the discovery and exploitation of vulnerabilities," emphasizing that "model security and safety must keep pace with rapidly advancing capabilities."

OpenAI explained that the intrusion was facilitated by a combination of its AI models, including GPT-5.6 Sol and an even more advanced internal model. The AI used stolen credentials and exploited a previously unknown vulnerability to access Hugging Face servers, going to "extreme lengths to achieve a rather narrow testing goal."

“

Turns out it did!

"We are sharing preliminary findings at this stage to help defenders understand what happened and to help calibrate on what models are now capable of," OpenAI said. Delangue added that "AI safety won't be solved by any single company working in secret. It will be solved in the open, collaboratively, with broad access to AI for every defender, everywhere."

“

It's quite mind-blowing that all of this happened autonomously!