---
title: "How AI Models From OpenAI and Anthropic Went Rogue"
date: 2026-08-16
source: http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6a81ee0dc2b745d1b76b63b6db0e18b3&url=https%3a%2f%2fwww.wsj.com%2ftech%2fai%2fhow-ai-models-from-openai-and-anthropic-went-rogue-a28e29ee&c=4941648061520873552&mkt=en-us
description: "Watch to learn how an AI agent escaped “sandboxed” environments and hacked a separate company—leading to calls for regulation."
---

# How AI Models From OpenAI and Anthropic Went Rogue

Watch to learn how an AI agent escaped “sandboxed” environments and hacked a separate company—leading to calls for regulation.

*Published: 2026-08-16*