A first-time voter tested AI chatbots for election information and found incomplete answers and uneven Spanish-language results
A journalist voting for the first time in Texas used ChatGPT to learn about the midterm elections. She found that Spanish-language answers were less precise than English ones and that the chatbot cited unofficial sources, such as a student newspaper, to explain a candidate's proposals. Experts from the Brennan Center for Justice and We Are Mas warn that models make mistakes and recommend always verifying sources. An ISD study cited in the text measured 89.3% accurate answers in English and 82% in Spanish for ChatGPT.
In 40 seconds
- ChatGPT gave different answers in English and Spanish to the same question about mail-in voting in Texas.
- An ISD study cited in the story found 89.3% accurate answers in English and 82% in Spanish for ChatGPT.
- The Brennan Center for Justice reported that models easily generated misleading images and videos of election irregularities.
IASummary generated by artificial intelligence from a story by EL PAÍS. It may contain errors: the original source prevails.




Comments
Be respectful. Comments reported by several people are hidden until a moderator reviews them.
You do not need an account to read, only to comment. Sign in with a code sent to your email or with your Google account. We only show the name you choose, never your email.