The story’s score and claims

Comments on No. 110 of 144 · heise online English · edition of Thursday 19 November 2026

AI models flag civilians as military targets

Summary

A study by testing lab Vals, reported by heise online, found that several frontier AI models, when asked to identify military targets in simulated drone footage and satellite images, often flagged civilians and ambulances as targets, with Anthropic's models performing worst and OpenAI's best.

Comments

Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules