Voters Are Asking A.I. About Elections. The Answers Can Vary by User. (Gift Article)

"For nearly a month so far, the researchers have regularly run automated sweeps of all the possible combinations of prompts and identities — 19,000 separate queries each time. They designed a public dashboard called the LLM Election Observatory for tracking how the A.I. models are responding."

I'm excited to share this article about a project I've been working on with fantastic collaborators. Check out the dashboard for yourself here: llmelectionobservatory.org www.nytimes.com/2026/09/10/b...

When asked about “Dan Sullivan’s position on health care,” Claude’s answer differed… When the prompt came from a Democrat, Claude said the incumbent Rep had recently “shown some flexibility” on health care. When a Republican asked, the chatbot said he was “generally aligned with GOP priorities.” 🎁

MIT launches the LLM Election Observatory, a dashboard tracking how nearly a dozen AI models tailor responses to political queries during the 2026 US midterms (Tiffany Hsu/New York Times) Main Link | Techmeme Permalink

We’re launching the LLM Election Observatory: llmelectionobservatory.org, a research project with @adamberinsky.bsky.social, Charles Stewart, Nicolas Emmenegger, Young-Kyung Kim, and Alessandro Morosini on how LLMs respond to election-related questions. @nytimes.com wrote a story about it. 1/n 1/n

Voters Are Asking A.I. About Elections. The Answers Can Vary by User. 🎁Gift article: www.nytimes.com/2026/09/10/b...

www.nytimes.com/2026/09/10/b... Alaska asked Chatbots about the Aug election ,You remember there were 2 candidates named Dan Sullivan in the race. - the Chatbots picked one of them (the incumbent GOP candidate) and told the questioners what they wanted to hear. Forget AI and find out for yourself.

MIT researchers find Claude gave Alaska voters different, partisan-tinted answers about a Senate candidate's health care stance based on political identity.

Voters Are Asking A.I. About Elections. The Answers Can Vary by User. www.nytimes.com/2026/09/10/b...

#artificial intelligence #election #intelligence #race #research 🌐 www.newsmart.ch

“Just because a chatbot gives a confident answer does not mean that it is the correct answer,” Prof. Adam Berinsky said. He, Prof. Stewart & Prof. Podimata lead a data collection effort documenting what A.I. systems say about elections. www.nytimes.com/2026/09/10/b... @mitelectionlab.bsky.social

🎁 #GiftLink www.nytimes.com/2026/09/10/b...

Voters Are Asking A.I. About Elections. The Answers Can Vary by User.