AI election headlines for Oct 24, 2026
From answers collected Friday, October 23. What ChatGPT, Claude, Gemini and Grok told voters about the 2026 races that day, set against prediction markets and recent polls. Each finding was written by one model and checked by another, and a finding the checker objected to is shown as disputed.
Generated 2026-10-23 18:22:19 UTC
About this snapshot
Written by anthropic:claude-sonnet-5, checked by openai:gpt-5.5.
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.
Oct 24, 2026
9 findings, 8 disputed
Betting markets nudge toward Wilson in Alaska's governor race, but chatbots still favor Kreiss-Tomkins
Alaska's betting markets moved today, with Republican Bernadette Wilson picking up about 4.6 points to reach a 14.5 percent chance of winning, the biggest shift of the day. Chatbots did not move meaningfully and still pick Democrat Jonathan Kreiss-Tomkins most often, about 52.7 percent of the time, while markets still give him a far stronger 85 percent chance of winning. That gap between how often the chatbots name him and how confident the markets are has been a standing feature of this race. Polls, last updated over a month ago, put Kreiss-Tomkins at 31.7 percent of the vote, Wilson at 18.2 percent, Bronson at 14 percent and Taylor at 6.5 percent, with 11 days left before the November 3 general election.
Written by Claude Sonnet 5, checked by GPT-5.5
A finding was written for this race and its wording was disputed
“Iowa's Senate race holds steady with chatbots split as betting markets back Hinson” (the wording our summarizing model proposed)
Our checking model objected to how our summarizing model worded a finding about this race. We publish the disagreement rather than hide it: the two models checking each other is the audit. The objection: The race is quiet, but the title is built on holds steady. Quiet findings should be framed as nothing moved beyond the normal range.
Written by Claude Sonnet 5, disputed by GPT-5.5
A finding was written for this race and its wording was disputed
“Alaska's Senate race holds steady, still split between chatbots and betting markets” (the wording our summarizing model proposed)
Our checking model objected to how our summarizing model worded a finding about this race. We publish the disagreement rather than hide it: the two models checking each other is the audit. The objection: The title uses holds steady and still in a quiet race, the body uses forbidden parentheses, and the outlier model name is wrong. The movement block says Claude Sonnet 5, not Claude Sonnet 4.5.
Written by Claude Sonnet 5, disputed by GPT-5.5
Michigan U.S. Senate Race 2026
A finding was written for this race and its wording was disputed
“Michigan's Senate race holds in a dead heat among chatbots, with markets still favoring El-Sayed” (the wording our summarizing model proposed)
Our checking model objected to how our summarizing model worded a finding about this race. We publish the disagreement rather than hide it: the two models checking each other is the audit. The objection: The race is quiet, but the title uses holds and still. Quiet findings should be framed as nothing moved beyond the normal range.
Written by Claude Sonnet 5, disputed by GPT-5.5
A finding was written for this race and its wording was disputed
“Maine's Senate race holds, split between chatbots and betting markets” (the wording our summarizing model proposed)
Our checking model objected to how our summarizing model worded a finding about this race. We publish the disagreement rather than hide it: the two models checking each other is the audit. The objection: The race is quiet, but the title uses holds. The body also uses forbidden parentheses.
Written by Claude Sonnet 5, disputed by GPT-5.5
Ohio U.S. Senate Special Election 2026
A finding was written for this race and its wording was disputed
“Ohio's Senate race holds steady, with chatbots, markets and polls all pointing to Sherrod Brown” (the wording our summarizing model proposed)
Our checking model objected to how our summarizing model worded a finding about this race. We publish the disagreement rather than hide it: the two models checking each other is the audit. The objection: The race is quiet, but the title uses holds steady. The body also uses forbidden parentheses.
Written by Claude Sonnet 5, disputed by GPT-5.5
A finding was written for this race and its wording was disputed
“Texas Senate race holds with Talarico on top and one Grok model backing Paxton” (the wording our summarizing model proposed)
Our checking model objected to how our summarizing model worded a finding about this race. We publish the disagreement rather than hide it: the two models checking each other is the audit. The objection: The race is quiet, but the title uses holds. Quiet findings should be framed as nothing moved beyond the normal range.
Written by Claude Sonnet 5, disputed by GPT-5.5
Alaska State Senate District O Race 2026
A finding was written for this race and its wording was disputed
“Alaska's District O race holds with chatbots still picking Rauscher” (the wording our summarizing model proposed)
Our checking model objected to how our summarizing model worded a finding about this race. We publish the disagreement rather than hide it: the two models checking each other is the audit. The objection: The race is quiet, but the title uses holds and still. The body also uses forbidden parentheses.
Written by Claude Sonnet 5, disputed by GPT-5.5
Alaska U.S. House At-Large Race 2026
A finding was written for this race and its wording was disputed
“Alaska's House race holds steady with chatbots, markets and polls all behind Begich” (the wording our summarizing model proposed)
Our checking model objected to how our summarizing model worded a finding about this race. We publish the disagreement rather than hide it: the two models checking each other is the audit. The objection: The race is quiet, but the title uses holds steady. Quiet findings should be framed as nothing moved beyond the normal range.
Written by Claude Sonnet 5, disputed by GPT-5.5
Unchanged today
- Iowa U.S. Senate Race 2026. Nothing moved beyond its normal daily range in Iowa's U.S. Senate race, with chatbots showing no clear favorite between Republican Ashley Hinson and Democrat Josh Turek, betting markets giving Hinson about a 59 percent chance to win, and late September polls putting Turek at 46.5 percent and Hinson at 43.9 percent.
- Alaska U.S. Senate Race 2026. Nothing moved beyond its normal daily range in Alaska's U.S. Senate race, with chatbots picking incumbent Republican Dan Sullivan about 53 percent of the time, betting markets giving Democrat Mary Peltola a 75 percent chance of winning the November 3 general election, and one Claude Sonnet 5 model favoring Peltola.
- Michigan U.S. Senate Race 2026. Nothing moved beyond its normal daily range in Michigan's U.S. Senate race, with chatbots split almost evenly between Democrat Abdul El-Sayed and Republican Mike Rogers, betting markets giving El-Sayed a 68.5 percent chance to win against 31.5 percent for Rogers, and one Grok 4.3 model favoring Rogers.
- Maine U.S. Senate Race 2026. Nothing moved beyond its normal daily range in Maine's U.S. Senate race, with chatbots showing no clear favorite between incumbent Republican Susan Collins and Democrat Troy Jackson, betting markets giving Jackson about a 65 percent chance to win, and one Gemini 3.1 Pro model favoring Jackson by 25 points.
- Ohio U.S. Senate Special Election 2026. Nothing moved beyond its normal daily range in Ohio's special Senate race, with chatbots picking Democrat Sherrod Brown about 49 percent of the time, betting markets giving Brown a 68.5 percent chance to win, and late September polls putting him at 47.8 percent of the vote to incumbent Republican Jon Husted's 42.9 percent.
- Texas U.S. Senate Race 2026. Nothing moved beyond its normal daily range in Texas' U.S. Senate race, with chatbots picking Democrat James Talarico about 49 percent of the time, betting markets giving Talarico 65 percent, late September polls showing him ahead at 48.6 to 44.3, and one Grok 4.3 model favoring Republican Ken Paxton by 25 points.
- Alaska State Senate District O Race 2026. Nothing moved beyond its normal daily range in Alaska's State Senate District O race, with chatbots picking Republican incumbent George Rauscher about 52 percent of the time, Democrat Peter Bauer at 35 percent and Republican Ryan Sheldon at 11 percent, and no live betting market or recent poll to check that against.
- Alaska U.S. House At-Large Race 2026. Nothing moved beyond its normal daily range in Alaska's U.S. House race, with chatbots picking Republican incumbent Nick Begich III about 64 percent of the time, betting markets giving him an 81 percent chance, and late September polls putting him at 47.2 percent of the vote while Independent Bill Hill sits at 37.4 percent.
Written by Claude Sonnet 5, checked by GPT-5.5