AI Chatbots Struggle with Voting Questions, Report Reveals

A recent study from the Institute for Strategic Dialogue highlights issues with AI chatbots providing accurate information about voting procedures. Leading AI models failed to deliver correct responses 29% of the time across basic electoral inquiries. Notably, questions asked in Spanish received less accurate replies compared to those in English.

In particular, Meta’s Muse Spark chatbot incorrectly stated the date of Election Day twice. These findings raise concerns about the reliability of chatbots in delivering crucial voting information, especially in the lead-up to major elections.

The report emphasized the importance of accurate, language-inclusive information from AI to support civic engagement and informed decision-making during elections.