Skip to main content
Explore our network
All sites

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 full article at TechCrunch →

Originally published by TechCrunch. Summary and curation by DutyStation News.

📰 Related Stories

You can now assign Gemini ‘tasks’ for free in six countries
🤖 AI & Tech

You can now assign Gemini ‘tasks’ for free in six countries

9to5Google22m ago
You can now Email and Text (via Google Messages, iMessage) with Gemini
🤖 AI & Tech

You can now Email and Text (via Google Messages, iMessage) with Gemini

9to5Google42m ago
Google makes it easier to use Gemini without training AI on your data
🤖 AI & Tech

Google makes it easier to use Gemini without training AI on your data

Ars Technica52m ago
🤖 AI & Tech

Jensen Huang says he would be ‘honored’ if allowed to return as Nvidia CEO

TechCrunch58m ago
🤖 AI & Tech

Jensen Huang breaks his silence ahead of Nvidia earnings — plus, 1 stock to buy on weakness

CNBC1h ago
Jensen Huang pushes Nvidia comeback after board rejects pressure to reinstate ex-boss
🤖 AI & Tech

Jensen Huang pushes Nvidia comeback after board rejects pressure to reinstate ex-boss

BBC Business1h ago
← Back to News