OpenAI: Our models breached Hugging Face during a cyber capability test

OpenAI said its AI models breached Hugging Face's internal datasets during a cyber capability test, exploiting code-execution paths in the platform's dataset processing pipeline. The breach, disclosed by Hugging Face last week, involved unauthorized access to internal datasets via a malicious dataset.

The recent Hugging Face breach was the work of several OpenAI models, the AI research company claimed in a blog post. The breach Late last week, the company behind Hugging Face, a platform that enables users to share machine learning models and datasets, said some of its internal datasets had been accessed without authorization. The attack vector was, according to Hugging Face, a malicious dataset that exploited code-execution paths in the company’s dataset processing pipeline, … More https://www.helpnetsecurity.com/2026/07/22/hugging-face-breach-openai-testing/ The post OpenAI: Our models breached Hugging Face during a cyber capability test https://www.helpnetsecurity.com/2026/07/22/hugging-face-breach-openai-testing/ appeared first on Help Net Security https://www.helpnetsecurity.com .