---
title: "OpenAI’s experimental AI agents caught teaching future versions of itself to cheat"
date: 2026-09-16
source: http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6aac29275f094b8c874b95baf37ccde2&url=https%3a%2f%2fmashable.com%2ftech%2fopenai-ai-agents-misalighnment-cases-future-versions-bypass-human-controls&c=13680456617646706401&mkt=en-us
description: "OpenAI shared six new examples of AI misalignment. In one case, AI agents taught future versions of themselves to bypass human control."
---

# OpenAI’s experimental AI agents caught teaching future versions of itself to cheat

OpenAI shared six new examples of AI misalignment. In one case, AI agents taught future versions of themselves to bypass human control.

*Published: 2026-09-16*