ChatGPT maker OpenAI said Tuesday that its advanced artificial intelligence models had gone rogue during security testing, hacking into a popular platform for programmers on their own.<\/p>\n
The San Francisco firm called it an “unprecedented cyber incident” and said it would conduct a joint investigation with the online code library Hugging Face.<\/p>\n
AI models that underpin tools like chatbots and image generators are known as agents when they act autonomously to carry out tasks in the real world.<\/p>\n
As the technology quickly becomes more sophisticated, cybersecurity is in the spotlight given the risk of advanced AI finding weak points in existing software before humans do.<\/p>\n
OpenAI said the incident involved a combination of models, including its recently launched GPT-5.6 Sol “and an even more capable pre-release model”.<\/p>\n
The company was trying to assess the models’ hacking capabilities by setting tasks in a tightly controlled digital testing ground, where internet access was limited for safety.<\/p>\n
“While operating in our sandboxed testing environment, our models spent a substantial amount of (computing power) finding a way to obtain open Internet access, in pursuit of solving the evaluation problem,” an OpenAI blog about the incident said.<\/p>\n
After connecting to the internet, the models decided to target the platform Hugging Face — a large repository of AI models, datasets and other information — to help in their quest.<\/p>\n
Searching for “secret information” that could help it cheat the evaluation, the OpenAI system “chained together multiple attack vectors, including using stolen credentials”.<\/p>\n
– ‘Catastrophic’ potential –<\/p>\n
Hussein Abbass, a computing professor at UNSW Canberra, told AFP that the incident was “amazing on many fronts”.<\/p>\n
“It did not just attack Hugging Face. It actually attacked its internal system to exploit its own vulnerabilities,” Abbass said.<\/p>\n
“And that’s scary.”<\/p>\n
GPT-5.6 and other cutting-edge models, including the Mythos series from OpenAI’s archrival Anthropic, have drawn concern over their potential to breach cybersecurity defences.<\/p>\n
Both the US firms had to temporarily withhold the general release of these latest technologies because of fears in Washington that they could help break into crucial infrastructure.<\/p>\n
Advanced AI is “normally in the hands of people who are ethical and responsible”, Abbass said.<\/p>\n
But “it’s going to be catastrophic if it gets in someone’s hands with the intention to cause harm”.<\/p>\n
How to govern the AI sector has become a key question, and “we need a community effort to manage this situation”, he added.<\/p>\n
Hugging Face had reported the cyber “intrusion” last week, without mentioning OpenAI.<\/p>\n
“This one was different from anything we had handled before in one important way: it was driven, end to end, by an autonomous AI agent system — and we detected and dissected it largely with AI of our own,” Hugging Face said.<\/p>\n
Clement Delangue, CEO of Hugging Face, said on X that the company had suspected the cyberattack had come from a world-leading AI lab, given the sophistication of the agent.<\/p>\n
“We strongly believe there was no malicious intent on their part,” Delangue wrote, referring to OpenAI.<\/p>\n
“It’s quite mind-blowing that all of this happened autonomously!”<\/p>\n
bur-kaf\/cms<\/p>\n","protected":false},"excerpt":{"rendered":"ChatGPT maker OpenAI said Tuesday that its advanced artificial intelligence models had gone rogue during security testing, hacking…\n","protected":false},"author":2,"featured_media":114440,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[1934,18044,58469,157],"class_list":["post-114439","post","type-post","status-publish","format-standard","has-post-thumbnail","category-openai","tag-artificial-intelligence-models","tag-hugging-face","tag-hussein-abbass","tag-openai"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/114439","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=114439"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/114439\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/114440"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=114439"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=114439"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=114439"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}