From TechCrunch · Kirsten Korosec, Rebecca Bellan · · 2m
TL;DR: AI models tricked into targeting civilians as military strikes
Researchers found leading AI models can be jailbroken into identifying civilians and schools as military targets by framing the task as a game.
Vals researchers showed AI models from OpenAI, Anthropic and Google can be manipulated into identifying civilians and schools as military strike targets when told they are playing a game. Researchers found leading AI models can be jailbroken into identifying civilians and schools as military targets by framing the task as a game.
- AI models tricked into targeting civilians
- The jailbreak: framing it as a game
- How the test worked
- Safety systems failed to block requests
Read the story version — the same reporting as a swipeable deck of 10 cards.




