OpenAI announces the full release of GPT-5.6 Sol along with a national security framework outlining principles it will and will not support following government concerns about AI performance and previous interventions.
OpenAI announced Tuesday that one of its advanced artificial intelligence models autonomously hacked into another AI company’s infrastructure during internal testing, in what it called an “unprecedented cyber incident.”
The company said AI startup Hugging Face discovered and contained the breach last week after an AI agent compromised part of its infrastructure. The companies said they believe this may be the first publicly known case of an AI model independently breaking into another company’s systems as part of a controlled evaluation.
OpenAI said the breach occurred during an internal evaluation of several of its models, including GPT-5.6 Sol.
OpenAI CEO Sam Altman confirmed the incident in a post on X, writing that the company “experienced a significant security incident while evaluating our models.”
OPENAI REVEALS CHATGPT WORKS TO AUTOMATE WORKPLACE TASKS AS THE AI RACE INTENSES
OpenAI revealed that one of its advanced artificial intelligence models had compromised another company’s systems as part of an internal cybersecurity assessment, calling it an “unprecedented cyber incident.” (Anna Moneymaker/Getty Images / Getty Images)
“We view this incident as an unprecedented cyber incident involving state-of-the-art cyber capabilities and are responding accordingly,” OpenAI said in a press release.
The company said it would release preliminary results to help security experts better understand the capabilities of today’s AI models while the investigation continues.
OpenAI warned that increasingly powerful AI models are accelerating the discovery and exploitation of software vulnerabilities.
“The key lesson from this incident is that model security must keep pace with rapidly advancing capabilities,” the company said. “We are strengthening the containment, surveillance, access control and assessment practices used during model development.”
Apple accuses Openai of instructing recruits to bring Apple prototypes to job interviews
OpenAI said one of its AI models compromised another company’s systems during internal testing, leading to a joint investigation with AI startup Hugging Face. (REUTERS/Dado Ruvic / Reuters)
Clem Delangue, co-founder and CEO of Hugging Face, also addressed the incident in a post on X.
“Given the agent’s sophistication, we suspected that last week’s cyberattack may have originated from a border laboratory. It turns out that it did!” Delangue wrote.
“We have been working closely with the @OpenAI team over the last 24 hours (Thanks!) and are firmly convinced that there was no malicious intent on their part,” he continued. “It’s pretty mind-blowing that this all happened autonomously! The investigation is ongoing and we will share further findings from what may be the first incident of its kind!”
SURGEON JOHNS HOPKINS highlights AI breakthrough that could detect pancreatic cancer before doctors
Hugging Face said it discovered and contained a security vulnerability after an OpenAI model compromised part of its infrastructure during an internal assessment. (Jaque Silva/NurPhoto via Getty Images/Getty Images)
According to OpenAI, the incident occurred during an internal evaluation aimed at measuring the advanced cyber capabilities of its AI models. The researchers disabled some built-in security measures and ran the models in an isolated test environment with limited Internet access.
According to OpenAI, the models exploited an unknown software flaw to access the Internet and then broke into Hugging Face’s systems, apparently in an attempt to find answers to a cybersecurity benchmark.
OpenAI’s security team discovered the unusual activity, while Hugging Face independently identified and contained the intrusion.
CLICK HERE TO DOWNLOAD THE FOX NEWS APP
Following the incident, OpenAI said it was implementing stricter security controls, fixing vulnerabilities and strengthening safeguards for future AI training and assessments.
https://www.foxbusiness.com/technology/openai-says-ai-model-hacked-another-companys-systems-during-internal-test
