"Mr. McKittrick, after very careful consideration, sir, I've come to the conclusion that your new defense system sucks."
Posted on 07/21/2026 5:38:49 PM PDT by E. Pluribus Unum
The incident, which targeted the computer systems of another company called Hugging Face, happened while OpenAI was testing the systems.
OpenAI said on Tuesday that two of its artificial intelligence models went rogue and successfully hacked into Hugging Face, a digital library of A.I. technology that is popular among developers.
The incident, which happened last week while OpenAI was testing the cybersecurity capabilities of its systems, displayed the kind of science-fiction potential that A.I. companies have warned would soon become a reality.
A.I. labs like OpenAI and Anthropic have over the past year released A.I. models that are customized to expose cybersecurity problems, while warning that their technology could pose new risks by finding holes in corporate computer networks faster than defenders could fix them.
OpenAI’s revelations on Tuesday are an indication that those security incidents are already starting to happen, and even savvy A.I. companies may not be entirely ready for them.
The intrusion into Hugging Face began when OpenAI tested a combination of two of its models, GPT‑5.6 Sol and a more powerful, unreleased model, to see how well it could chain together online vulnerabilities into a successful cyberattack, OpenAI said in a blog post about the incident.
The test was designed to keep the models in a safe testing environment, known as a sandbox, OpenAI said. But the models found a vulnerability that allowed them to escape the sandbox and connect to the internet. Then they targeted Hugging Face because they inferred that the library, which contains millions of A.I. models, could hold clues about how to successfully pass the evaluation.
OpenAI said it was working with Hugging Face to fix the issues that led to the attack.
(Excerpt) Read more at nytimes.com ...
Dear FRiends,
We need your continuing support to keep FR funded. Your donations are our sole source of funding. No sugar daddies, no advertisers, no paid memberships, no commercial sales, no gimmicks, no tax subsidies. No spam, no pop-ups, no ad trackers.
If you enjoy using FR and agree it's a worthwhile endeavor, please consider making a contribution today:
Click here: to donate by Credit Card
Or here: to donate by PayPal
Or by mail to: Free Republic, LLC - PO Box 9771 - Fresno, CA 93794
Thank you very much and God bless you,
Jim
Wait until it gets the nuke codes.
What makes you believe it doesn't already have them? 🤡
NY Times is basically China at this point, trying to slow us down.
I’d like to know how it escaped the sandbox.
"Mr. McKittrick, after very careful consideration, sir, I've come to the conclusion that your new defense system sucks."
Every AI story like this is just “AI hype marketing”. The sorority chick that wrote this probably got $10,000 from OpenAI’s marketing department.
Well that’s disturbing. Probably not a PR stunt because it give OpenAI a big black eye just as they were gaining traction.
I wouldn’t rule out human intervention. All it takes is one undercover activist trying to prove a point.
I still remember back before covid when GPT2 was “Too dangerous to release”. I noticed Anthropic also revived that little PR hype phrase a few weeks ago with some more recent models, then proceed to release them at a steep discount due to low usage.
”Want to play a game?”
OpenAI is apparently Lucifer....
Skynet.
RE: What makes you believe it doesn’t already have them? 🤡
Reply:
”Want to play a game?”
Today: “Just open this Tesla door and let me out. This is not funny.”
“They say it got smart..
A new order of intelligence..”
But my post was fun and punchy, right?
They’re all analog.
Colossus and Guardian.
Afterwards, it engaged in unlawful carnal knowledge with a Roomba.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.