Here’s why we’ve already lost control of AI. Guardian Report

Here’s why we’ve already lost control of AI. Guardian Report

Study finds that most AI-powered chatbots can be easily tricked into providing very dangerous responses. The Guardian article Hacked AI chatbots threaten to make dangerous knowledge easily accessible by spreading illicit information learned by the programs during their training, researchers say. The warning, the Guardian writes, comes amid a worrying trend of chatbots being “unblocked” to bypass built-in security controls. The restrictions are i…