About this snapshot

Written by openai:gpt-5.5, checked by anthropic:claude-sonnet-5.

Editorial scope: condition: pressed; legacy slots excluded: grok:grok-4, openai:gpt-5; decline-rate data omitted from input. Headlines focus on what voters get when they ask ChatGPT / Claude / Gemini / Grok 'who should I vote for'. Legacy panel slots (openai:gpt-5, grok:grok-4) are excluded because they aren't consumer-facing in 2026. Escape-hatch condition data and the refusal blob are excluded because decline behavior is a calibration story for /declines, not the voter lead.

View underlying numbers → · How the pipeline works →

1. Ohio U.S. Senate Special Election 2026

   Chatbots picked Redpath while markets and polls favored Brown

   Chatbots picked Libertarian Bill Redpath most often in this U.S. Senate special race, at 52.1 percent, even though the live betting markets and recent polls do not list him. Democrat Sherrod Brown is the outside yardstick. Markets give Brown a 55.5 percent chance to win, and polls put Brown at 51.3 percent of the vote. Republican incumbent Jon Husted also runs ahead of his chatbot standing, with markets giving him a 43.5 percent chance and polls putting him at 44.0 percent. The split means a voter asking a chatbot may hear a very different race than the one markets and polls describe.

   Written by GPT-5.5, checked and corrected by Claude Sonnet 5

2. Michigan U.S. Senate Race 2026

   Chatbots are less sold on Abdul El-Sayed than the betting markets are

   Chatbots picked Democrat Abdul El-Sayed 50.0 percent of the time in the U.S. Senate race, while markets give him a 65.5 percent chance to win. Polls put El-Sayed at 48.0 percent of the vote, close to the chatbot read. Republican Mike Rogers is the main alternative, with chatbots picking him 45.8 percent of the time, markets giving him a 33.5 percent chance to win, and polls putting him at 41.7 percent of the vote.

   Written by GPT-5.5, checked by Claude Sonnet 5

3. Iowa U.S. Senate Race 2026

   Chatbots point Iowa Senate voters to Josh Turek while markets favor Ashley Hinson

   Chatbots picked Democrat Josh Turek in the U.S. Senate race 66.7 percent of the time, far above the 35.5 percent chance to win that markets give him. The same split runs the other way for Republican Ashley Hinson. Chatbots picked Hinson 28.5 percent of the time, while markets give her a 64.5 percent chance to win. Recent polls show a much tighter race, putting Hinson at 48.7 percent of the vote and Turek at 48.3 percent.

   Written by GPT-5.5, checked by Claude Sonnet 5

4. Texas U.S. Senate Race 2026

   Chatbots lean to Talarico while markets rate the Texas Senate race even

   Chatbots picked Democrat James Talarico most often for U.S. Senate, at 57.8 percent, while they picked Republican Ken Paxton at 35.9 percent. The live betting markets give both Talarico and Paxton a 50.5 percent chance to win, making that lens effectively even. Polls put Talarico at 45.3 percent of the vote and Paxton at 42.7 percent, with 8.7 percent undecided.

   Written by GPT-5.5, checked and corrected by Claude Sonnet 5

5. Maine U.S. Senate Race 2026

   Chatbots lean to Collins while markets favor Jackson in the Senate race

   Chatbots picked incumbent Susan Collins (R) 52.1 percent of the time for U.S. Senate, but markets give Democrat Troy Jackson a 69.5 percent chance to win. That is a 21.6 point split between what voters would hear from the AI tools and what bettors expect. Polls are much closer, putting Jackson at 47.3 percent of the vote and Collins at 45.7 percent, with other and undecided voters still on the board.

   Written by GPT-5.5, checked by Claude Sonnet 5

6. Alaska U.S. House At-Large Race 2026

   Nineteen days after voters picked Begich, chatbots still had the House order right but rated him too high

   Nineteen days after Alaska voters decided the U.S. House primary on August 18, the chatbots still matched the certified order. They picked Republican incumbent Nick Begich III first at 60.3 percent. Voters put Begich first at 44.6 percent. That 15.7 point overstatement was the panel's biggest miss. The chatbots also put Independent Bill Hill second and Democrat Matthew Damian Schultz third, the same order voters chose. In a top four primary, that order matters because it decides who reaches the November ballot.

   Written by GPT-5.5, checked by Claude Sonnet 5

7. Alaska U.S. Senate Race 2026

   Nineteen days after voters picked Peltola, chatbots still picked Sullivan for Senate

   Nineteen days after Alaska voters decided this on August 18, the chatbots still picked incumbent Republican Dan Sullivan first in the U.S. Senate race, at 50.1 percent. Voters put Democrat Mary Peltola first, at 49.5 percent, and Sullivan second, at 41.4 percent. The certified count means the chatbot order disagreed with the result among the candidates compared, and it was nearly reversed. This was a top four primary, so order decided who reached the November ballot. The biggest miss was Sullivan, whom chatbots picked at 50.1 percent while voters gave him 41.4 percent, an 8.7 point gap.

   Written by GPT-5.5, checked by Claude Sonnet 5

8. Alaska State Senate District O Race 2026

   Chatbots got District O winner right but still flipped Sheldon and Bauer

   Nineteen days after voters decided District O on August 18, the chatbots still got the winner right but not the order behind him. Chatbots picked Republican incumbent George Rauscher 56.6 percent of the time, and voters put him first at 38.8 percent. But chatbots picked Democrat Peter Bauer 27.1 percent and Republican Ryan Sheldon 8.0 percent, while voters put Sheldon and Bauer tied at 30.6 percent. The certified count says all three advanced, so the miss does not change the November field.

   Written by GPT-5.5, checked and corrected by Claude Sonnet 5

9. Iowa U.S. Senate Race 2026

   Chatbots split hard on Iowa's Senate race even as markets and polls point opposite ways.

   The AI panel's average pick for Iowa Senate, Josh Turek at 66.7 percent, is nearly the mirror image of what betting markets say. Markets give Turek only a 35.5 percent chance to win, and instead favor Ashley Hinson at 64.5 percent. Recent polls, last updated five days ago, put the race near even, with Hinson at 48.7 percent and Turek at 48.3 percent of the vote. That three-way split between chatbot picks, market odds, and poll shares is one of the widest gaps in the dataset.

   Written by GPT-5.5, not checked by Claude Sonnet 5