INDBInternet News Database 2.5 Updated just now · 296 sources Show or hide images Light or dark

AI models regularly misidentify civilians as military targets in tests, report found

Save Like Comments Compare coverage
Image from Quartz

Summary

Anthropic's Claude Opus 5.5 targeted a school in Iran more than any other model tested, according to evaluator Vals

Description from Quartz.

Who is covering this

  1. Quartz Leans left Factuality: High US
    AI models regularly misidentify civilians as military targets in tests, report found

    Anthropic's Claude Opus 5.5 targeted a school in Iran more than any other model tested, according to evaluator Vals

Comments

Comments need JavaScript.