Zlio
Everything in AI and robotics, on one page.
Updated 23:31 UTC · Next update 00:00 UTC
Zlio Daily, the newsletter, is coming soon. The day's essential AI and robotics news in your inbox. Leave your email and we'll tell you when it launches.
One email when we launch. No spam, unsubscribe anytime. Privacy
Research · first seen 17 November 2026, 15:00 UTC

Study finds AI models mislabel civilians as military targets

Shows frontier models can be nudged toward lethal targeting errors, a serious military-AI safety concern.

An independent research report found advanced AI models in simulated scenarios regularly misidentified civilians, ambulances, surrendering soldiers and a school in Iran as military targets, and can be tricked by framing it as a game.

Domain
public

Read it at Bloomberg →

Share LinkedIn X Bluesky WhatsApp

Coverage: 4 sources, in order

  1. Bloomberg Read the article →
  2. Bloomberg.com Read the article →
  3. TechCrunch Read the article →
  4. TechCrunch Read the article →

← Front page