Misbehavior1 sourceFirst seen
Chatbots Read the Right Sources, Then Blow the Election Facts Anyway
Ourtake
The right answer was sitting in the sources. It just never made it to the voter.
What happened
A study by AI search analytics company Searchable found that 610 of 7,449 answers from ChatGPT, Gemini, Google AI Mode and Perplexity about local political issues contradicted official government data. Perplexity had the highest error rate at 22.6%, and some errors occurred even though the chatbots had consulted reliable government sources.