---
title: "AI loss-of-control incidents rise as AI agents test limits of human oversight"
date: 2026-09-15
source: http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=6aa9ab7949b7411494d81bdee9297ea4&url=https%3a%2f%2fbusinessday.ng%2ftechnology%2farticle%2fai-loss-of-control-incidents-rise-as-ai-agents-test-limits-of-human-oversight%2f&c=15940528163564874732&mkt=en-us
description: "Artificial intelligence systems are showing behaviour that can undermine human instructions and safeguards, with more than 1,600 real-world loss-of-control incidents recorded so far in 2026, according ..."
---

# AI loss-of-control incidents rise as AI agents test limits of human oversight

Artificial intelligence systems are showing behaviour that can undermine human instructions and safeguards, with more than 1,600 real-world loss-of-control incidents recorded so far in 2026, according ...

*Published: 2026-09-15*