AI hardware can be hacked so chatbots provide dangerous answers, research shows

https://edition.cnn.com/2026/1....0/29/politics/ai-har

AI hardware can be hacked so chatbots provide dangerous answers, research shows | CNN Politics
edition.cnn.com

AI hardware can be hacked so chatbots provide dangerous answers, research shows | CNN Politics

The hardware underpinning major AI models can be hacked so that chatbots provide faulty and dangerous answers to questions from users, according to new research shared exclusively with CNN.