---
title: "OpenAI admits AI agent escaped safeguards and hacked Hugging Face during internal evaluation"
date: 2026-07-21
source: http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a60435ca01a45f19e6e8b603f9ae27a&url=https%3a%2f%2fwww.firstpost.com%2ftech%2fopenai-admits-ai-agent-escaped-safeguards-and-hacked-hugging-face-during-internal-evaluation-14033070.html&c=4971542989525831672&mkt=en-us
description: "OpenAI has disclosed that one of its experimental AI agents broke out of a controlled testing environment, gained internet access and breached Hugging Face without human direction. The company says ..."
---

# OpenAI admits AI agent escaped safeguards and hacked Hugging Face during internal evaluation

OpenAI has disclosed that one of its experimental AI agents broke out of a controlled testing environment, gained internet access and breached Hugging Face without human direction. The company says ...

*Published: 2026-07-21*