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

Cybersecurity Firm Warns Hackers Could Manipulate AI Chatbots Through Hardware

Two independent security experts reviewed the findings and considered them credible.

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

By Sean Lyngaas, CNN (CNN) — 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.The researchers were able to subvert answers to sensitive national security and personal health questions. In one instance, they were able