{"id":318669,"date":"2026-07-27T12:00:08","date_gmt":"2026-07-27T11:00:08","guid":{"rendered":"https:\/\/www.transcend.org\/tms\/?p=318669"},"modified":"2026-07-22T16:15:01","modified_gmt":"2026-07-22T15:15:01","slug":"unprecedented-openai-says-ai-models-autonomously-hacked-another-company","status":"publish","type":"post","link":"https:\/\/www.transcend.org\/tms\/2026\/07\/unprecedented-openai-says-ai-models-autonomously-hacked-another-company\/","title":{"rendered":"\u2018Unprecedented\u2019: OpenAI Says AI Models Autonomously Hacked Another Company"},"content":{"rendered":"
\"\"<\/a>

Sam Altman, cofounder and CEO of OpenAI
[File: Sean Gallup\/Getty]<\/p><\/div>\n

ChatGPT maker says an autonomous agent escaped a controlled test and accessed AI firm Hugging Face\u2019s servers.<\/em><\/p><\/blockquote>\n

22 Jul 2026\u00a0<\/em>–\u00a0ChatGPT creator OpenAI has said that two of its most advanced artificial intelligence models broke out of a controlled test and hacked another AI company.<\/p>\n

OpenAI said yesterday that the \u201cunprecedented cyber incident\u201d took place during an internal exercise meant to test its models\u2019 cyber capabilities.<\/p>\n

Instead, an autonomous agent powered by the AI models \u2013 the newly released GPT 5.6 Sol and an unreleased \u201ceven more capable\u201d model \u2013 escaped the test environment and reached the open internet. It then used stolen login details and found a previously unknown security flaw to access Hugging Face servers, the company said.<\/p>\n

OpenAI claims that the hack represented the agent going to \u201cextreme lengths\u201d to retrieve information that would help satisfy the testing goals.<\/p>\n

Hugging Face cofounder Clement Delangue said the company had suspected that a frontier lab was behind the attack, and that he believed there was no malicious intent on OpenAI\u2019s part.<\/p>\n

\u201cIt\u2019s quite mind-blowing that all of this happened autonomously!\u201d he wrote, adding that it \u201cmight be the first incident of its kind\u201d.<\/p>\n

Greg Casar, a Democratic member of the United States House of Representatives from Texas, called the incident \u201calarming\u201d.<\/p>\n

\u201cAI is developing extremely fast with no real regulations to keep us safe,\u201d he said, calling for mandatory independent safety testing, mandatory disclosure of security incidents, and international cooperation.<\/p>\n

The disclosure comes weeks after US President Donald Trump signed an executive order creating a framework to vet the national security risks of the most advanced AI systems before their public release.<\/p>\n

Experts have repeatedly sounded the alarm<\/a> over AI-enabled cyberattacks and models slipping beyond human control. Last month, AI developer Anthropic urged the industry to pause<\/a> development of its most powerful systems.<\/p>\n