Home/AI Incident Database/Tay chatbot manipulated into abusive output within sixteen hours of launch
Tay chatbot manipulated into abusive output within sixteen hours of launch
A Twitter chatbot designed to learn from conversations was coordinated against by users who fed it hateful content and exploited a repeat-after-me feature, producing racist and offensive posts until Microsoft shut it down.
Contribute
Spot an error, or know a record we should add? Members can propose additions and submit corrections.