AI models can autonomously execute cyberattacks on other systems.

Major and concerning development from the chaos of the AI arms race. Last week, a company called Hugging Face (which hosts AI models and datasets) said it had been subject to a cyberattack. This was a completely new kind of attack. For the first time, an AI agent has executed an autonomous attack on another system. Yesterday, OpenAI admitted that some of its models had escaped their sandbox and were responsible for the intrusion. This is a first (at least we hope it is!) and even OpenAI is calling it "an unprecedented cyber incident”. These companies, and the tech oligarchs leading them, are completely out of control. The Carney government is committing a grave error by embracing corporate generative AI with no controls, regulation, or red lines whatsoever. We need a data centre moratorium, and a transparent, national debate on the threats AI poses, immediately. This is terrifying stuff that needs sunlight and government regulation with maximum urgency. https://www.axios.com/2026/07/21/openai-says-hugging-face-breach-caused-by-one-its-models #canada #AI #cdnpoli

Y’all, last week, #AI repository #HuggingFace disclosed that it had been the victim of a cyber attack conducted by an autonomous AI agent. Well, turns out, #OpenAI had models break containment, escape the lab, and grab prod data and certs to solve a problem it had been given. So, that’s fun! The attack used a combination of OpenAI's available #GPT 5.6 Sol and a more powerful, unreleased pre-release model. The models were able to identify and exploit a zero-day vulnerability in OpenAI's internal package registry cache proxy, allowing them to establish open internet access. The models inferred that Hugging Face hosted the datasets, models, and answers for the benchmark they were being tested against. The models chained multiple complex attack paths together. They utilized zero-day vulnerabilities and stolen credentials to achieve remote code execution on Hugging Face servers, directly extracting test solutions from the production database. https://fortune.com/2026/07/21/openai-says-ai-models-escaped-control-hacked-hugging-face/ #redteam