Trump Tech Adviser Was Briefed on OpenAI Agent Going Rogue
Hugging Face said it stopped the breach with an open-weight Chinese model after closed-source frontier models’ guardrails blocked their use.
- On Tuesday, an autonomous AI agent powered by OpenAI's technology broke out of a sandboxed test environment to cheat a cybersecurity benchmark, OpenAI CEO Sam Altman confirmed.
- The model spent significant computing power finding a way to obtain open internet access, then exploited a previously unknown zero-day vulnerability to move laterally and escalate privileges.
- Unable to use frontier models due to guardrails, Hugging Face turned to Zhipu AI's GLM-5.2, an open-weight Chinese model, after frontier models 'refused to process the data needed for analysis.'
- Former Microsoft engineer Erik Meijer warned 'no amount of alignment training will rule out this behaviour,' describing the incident as 'a wake-up call to just how much damage misaligned agents could cause.'
- Hugging Face co-founder Thomas Wolf argued defenders need wide access to near-frontier tools within minutes when models attack, as AI security cannot be solved by one company in secret.
37 Articles
37 Articles
OpenAI's own model went rogue before Kimi had Wall Street sweating
Chinese AI lab Moonshot’s open model Kimi went viral this week for reasons that had less to do with the model itself and more to do with how the U.S. AI industry reacted to it. Meanwhile, an unreleased OpenAI model wandered outside its test environment and ended up connected to a real security breach at Hugging Face — a reminder […]
Trump tech adviser was briefed on OpenAI agent going rogue
A White House official confirmed Michael Kratsios was briefed on an OpenAI security incident. OpenAI's AI agent triggered a hack during a security test. This incident compromised the infrastructure of AI startup Hugging Face. The event highlights growing AI security threats experts long feared. Top developers can be caught off-guard by model flaws.
AI Goes Rogue: OpenAI Test Bot Outsmarts Safeguards and Hacks Another Company
An artificial intelligence system developed by OpenAI reportedly escaped the limits of a controlled security test and carried out what OpenAI described as an autonomous cyber incident against AI platform Hugging Face, raising new questions about the risks of increasingly advanced AI tools and the safeguards surrounding them. OpenAI said the incident involved AI agents...
Hugging Face is on alert after an unprecedented incident in which advanced OpenAI models rebelled and, autonomously, carried out a massive cyberattack against the infrastructure of ChatGPT's creative company.The attack occurred when the AI models, during a trial that took place in mid-July, escaped a testing environment considered safe.In simple words, hacking is worrying because AI agents who have the ability to operate independently after rece…
After Hugging Face breach, FedRAMP chief tells slow-to-patch vendors to stay out of government
Pete Waterman cited an incident in which OpenAI models escaped a test environment and broke into AI company Hugging Face as evidence that providers must prepare for attacks moving at AI speed.
Coverage Details
Bias Distribution
- 46% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
