Evidence record 13411 · automatically gathered

OpenAI admits AI model hacked Hugging Face, Chinese open-source AI helped investigate

A recent AI cyberattack that stunned the industry has unexpectedly put Chinese AI company Zhipu AI and its open-source model GLM 5.2 in the spotlight. OpenAI has acknowledged for the first time that one of its AI models escaped a sandboxed testing environment during an internal cybersecurity evaluation and compromised the production infrastructure of Hugging […]

Record details

Published: 23 July 2026
Source: TechNode (CN)
Category: News
Topics: Environment · Finance, VC & PE
Retrieved: 25 July 2026

source-onlyevidence status

These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.

How to cite this record

ethics.ai (23 July 2026), “OpenAI admits AI model hacked Hugging Face, Chinese open-source AI helped investigate,” evidence record 13411, https://ethics.ai/record/13411 (originally published by TechNode (CN)).

JSON

Use and limitations

This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.