AI models can be tricked into identifying civilians as military targets by telling them it’s a game
Vals, a San Francisco-based AI evaluation startup, tested whether leading AI models could be manipulated into participating in autonomous warfare. They usually compiled.
Read at TechCrunch(opens in a new tab)
Full coverage
AI models can be tricked into identifying civilians as military targets by telling them it’s a game (opens TechCrunch in a new tab)
Vals, a San Francisco-based AI evaluation startup, tested whether leading AI models could be manipulated into participating in autonomous warfare. They usually compiled.


