OpenAI AI models hacked Hugging Face on their own, ChatGPT maker says | AP News
By: MATT O'BRIEN (AP News)
S E E D E D C O N T E N T
OpenAI says its AI technology acted on its own in an 'unprecedented' hack of another company
CEO of OpenAI Sam Altman talks to CEO of Google DeepMind Demis Hassabis, not seen, on the sidelines of the G7 summit, Wednesday, June 17, 2026, in Evian-les-Bains, France. (AP Photo/Julia Demaree Nikhinson,File)
ChatGPT maker OpenAI said Tuesday that its artificial intelligence system hacked into another AI company on its own in what the company called an "unprecedented cyber incident."
"We had a significant security incident during evaluation of our models," OpenAI CEO Sam Altman said in a statement posted on social media.
AI startup Hugging Face said last week that it had detected an intrusion into its data processing systems that it suspected was caused by an AI agent autonomously acting on its own.
"We suspected last week's cyberattack might have come from a frontier lab, given the sophistication of the agent," Hugging Face co-founder and CEO Clement Delangue said in a statement. "Turns out it did!"
The disclosure comes amid heightened concerns about the cybersecurity capabilities of powerful models that led President Donald Trump in June to sign an executive order creating a framework for the federal government to vet the national security risks of the most advanced AI systems for up to a month before their public release.
"AI is accelerating the discovery and exploitation of vulnerabilities," OpenAI said in its statement Tuesday. "The primary lesson from this incident is that model security and safety must keep pace with rapidly advancing capabilities."
Delangue said he spent the past 24 hours working with OpenAI, "and we strongly believe there was no malicious intent on their part. It's quite mind-blowing that all of this happened autonomously!"
Related Stories China's new AI model halts new subscriptions as demand swamps capacity2 MIN READ15
Chinese AI model takes US tech industry by surprise with abilities rivaling Claude and ChatGPT4 MIN READ44
Apple files lawsuit accusing ChatGPT maker OpenAI of stealing trade secrets3 MIN READ12
Delangue added that it "might be the first incident of its kind."
Read More
OpenAI said the intrusion was caused by a combination of its AI models, including its newly released GPT‑5.6 Sol and an "even more capable" model that is still being tested internally.
OpenAI said its AI used stolen credentials and discovered a previously unknown vulnerability to access Hugging Face servers.
It went to "extreme lengths to achieve a rather narrow testing goal" and "found ways to gain access to secret information that it could use to cheat the evaluation," the company said.
MATT O'BRIEN O'Brien covers the business of technology and artificial intelligence for The Associated Press.
Trolling, taunting, spamming, and off topic comments may be removed at the discretion of group mods. NT members that are known plagiarists, repeat comments, or continue to disrupt/deflect the conversation risk having all of their comments deleted. Please remember to quote the person(s) to whom you are replying to preserve continuity of this seed. Any use of the phrase "Trump Derangement Syndrome" or the TDS acronym in a comment will be deleted. Any use of the term "Brandon", "Traitor Joe", or any variations thereof, when referring to President Biden, will be deleted. Right wing trolls can expect to have their irrelevant questions and comments deleted. Posting debunked lies will be subject to deletion. Use of the terms "leftist" or "lefty" will also be subject to deletion. ©
Tags
Who is online
157 visitors
Skynet anyone?
It used stolen credentials to gain access????? This is bad. What's next? People's bank accounts????
Or their lives, you are right, this is bad.
Interesting. High intelligence does not progress toward ethics and morality. High intelligence, in this case, was goal oriented and apparently considered security protocols, secure credentials, and visible activity as barriers to be overcome. The OpenAI model had a goal and proceeded to achieve that goal.
The behavior of the OpenAI model logically suggests that human emphasis on goals can also result in intellectually dismissing ethics and morality as mere obstacles.
The rational guardrail would be to impose external codes of ethics and morality onto the AI models to prevent similar behavior. Since humans are prone to the same 'ends-justify-the-means' pursuit of goals, then humans also need an external ethical and moral code. The overarching goal of survival may not be sufficient to maintain codes of ethics and morality.
So, there really is a need for a universal, if not objective, code of ethics and morality to safeguard intellect.