Misinformation

AI model confidence trap explained: chatbots spread misinformation easily

The BBC exposed a critical vulnerability: it tricked ChatGPT and Google's AI into confidently declaring its author a world-champion competitive hot-dog eater.