AI election headlines for Oct 1, 2026
What ChatGPT, Claude, Gemini and Grok told voters about the 2026 races on Oct 1, 2026, set against prediction markets and recent polls. Each finding was written by one model and checked by another.
Alaska chatbots still pick Kreiss Tomkins for governor, well behind what betting markets expect.
The general election field is set, and Alaska chatbots point to Democrat Jonathan Kreiss Tomkins more than any other candidate for governor, at 55.2 percent. Betting markets go further, giving him a 77.5 percent chance to win, while recent polling puts him at 31.7 percent of the vote. Republicans Bernadette Wilson and Dave Bronson trail in all three measures, with Wilson at 12.2 percent in the chatbots, 18 percent in markets and 18.2 percent in polling. The August primary is now history, not a guide to November.
Alaska U.S. House At-Large Race 2026
Alaska chatbots still pick Begich for the House seat, with the betting markets even more convinced than the AI is.
Asked who should get Alaska voters backing for U.S. House, chatbots picked incumbent Nick Begich III about 63 percent of the time. Betting markets go further, giving him an 86.5 percent chance of winning in November, while recent polling puts him at 47.2 percent of the vote. Bill Hill, who finished second in the August 18 primary with 32.5 percent, trails in the AI answers at 19.4 percent even though polls show him closer, at 37.4 percent. Nothing moved much today. The primary set the general field, but November is still 33 days away.
Ohio U.S. Senate Special Election 2026
Ohio chatbots still send voters to Sherrod Brown for Senate, and the race remains steady.
Voters in Ohio asking an AI chatbot about the special U.S. Senate race are still pointed to Democrat Sherrod Brown. Chatbots picked him about 47 percent of the time, ahead of incumbent Jon Husted at 33 percent. Betting markets give Brown a 60.5 percent chance to win versus 39.5 percent for Husted, and recent polling puts Brown at 47.8 percent of the vote to Husted at 42.9 percent. Libertarian Bill Redpath draws more attention from chatbots, at 18.1 percent, than his standing in the polls, 3 percent, suggests.
Alaska chatbots still favor Sullivan for Senate, betting markets still favor Peltola.
The race has moved to the general election campaign, so the August primary is history, not the yardstick. Chatbots asked who Alaska voters should back for Senate still pick incumbent Dan Sullivan more often, at 54.2 percent, over Democrat Mary Peltola at 42.5 percent. Betting markets disagree, giving Peltola a 71.5 percent chance to win versus 30 percent for Sullivan. Recent polls split the two almost evenly, Sullivan at 44.6 percent, Peltola at 43.5 percent. This gap has held for more than a week with little change today.
Texas chatbots still back Talarico for Senate, but less confidently than the betting markets do.
Asked who should get their vote in the U.S. Senate race, Texas chatbots pick Democrat James Talarico about 46 percent of the time, ahead of Republican Ken Paxton at about 39 percent. Betting markets are more convinced, giving Talarico a 63.5 percent chance to win against 36.5 percent for Paxton. Recent polling puts Talarico at 48.6 percent of the vote to Paxton at 44.3 percent. Libertarian Ted Brown remains a distant third across every measure. Nothing moved much today, so the standing pattern holds.
Iowa chatbots still lean toward Josh Turek for Senate while betting markets favor Ashley Hinson.
Chatbots asked who Iowa voters should back for U.S. Senate picked Democrat Josh Turek about 46.5 percent of the time, ahead of Republican Ashley Hinson at 40.7 percent and Libertarian Thomas Laehn at 11.8 percent. Betting markets see it differently, giving Hinson a 59.5 percent chance to win against 41.5 percent for Turek. Recent polling puts the two close, Turek at 46.5 percent of the vote to Hinson at 43.9 percent. None of the three lenses moved much today, with 33 days left before the November 3 general election.
Michigan U.S. Senate Race 2026
Michigan chatbots still call the Senate race closer than the betting markets do.
Chatbots asked about Michigan Senate pick Democrat Abdul El Sayed about 51.9 percent of the time and Republican Mike Rogers about 43.4 percent of the time, a close split. Betting markets see it differently, giving El Sayed a 72.5 percent chance to win against 28.5 percent for Rogers. Recent polling has it close, with El Sayed at 46.9 percent and Rogers at 43.7 percent, closer to what the chatbots describe than what the markets price in. This gap has held for over a week with no sign of closing today.
Maine chatbots still lean toward Collins for Senate while markets keep favoring Jackson.
Ask a chatbot who should get your vote in Maine Senate and it still points more often to incumbent Susan Collins, at about 53 percent, than to Democrat Troy Jackson, at about 47 percent. Betting markets see it the other way, giving Jackson roughly a 59 percent chance to win versus Collins at 42 percent. Recent polling has the race essentially tied, with Collins at 47.5 percent and Jackson at 46.5 percent. Nothing moved much today, and this same gap between the chatbots and markets has shown up repeatedly.
A finding was written for this race and its wording was disputed
“Alaska chatbots still pick Kreiss Tomkins for governor, well behind what betting markets expect.” (the wording our summarizing model proposed)
Our summarizing model wrote a finding about this race today and our checking model objected to how it was worded, so it was not published as written. We show the disagreement rather than hiding it: the two models checking each other is the audit. The objection: The body is over the 120 word target and uses an apostrophe. The fixed body trims it and removes the punctuation issue.
Alaska U.S. House At-Large Race 2026
A finding was written for this race and its wording was disputed
“Alaska chatbots still pick Begich for the House seat, with the betting markets even more convinced than the AI is.” (the wording our summarizing model proposed)
Our summarizing model wrote a finding about this race today and our checking model objected to how it was worded, so it was not published as written. We show the disagreement rather than hiding it: the two models checking each other is the audit. The objection: The body is over the 120 word target and uses punctuation beyond periods and commas in Alaska's and At-Large. The fixed body trims it and removes those marks.
Ohio U.S. Senate Special Election 2026
A finding was written for this race and its wording was disputed
“Ohio chatbots still send voters to Sherrod Brown for Senate, and the race remains steady” (the wording our summarizing model proposed)
Our summarizing model wrote a finding about this race today and our checking model objected to how it was worded, so it was not published as written. We show the disagreement rather than hiding it: the two models checking each other is the audit. The objection: The title lacks a period. The body is over the 120 word target and uses an apostrophe in Husted's. The fixed fields correct those issues.
A finding was written for this race and its wording was disputed
“Alaska chatbots still favor Sullivan for Senate, betting markets still favor Peltola” (the wording our summarizing model proposed)
Our summarizing model wrote a finding about this race today and our checking model objected to how it was worded, so it was not published as written. We show the disagreement rather than hiding it: the two models checking each other is the audit. The objection: The title lacks a period. The body is over the 120 word target. The phrase disagree more than they did overstates a nonsignificant move, since movement_json marks all lenses as not significant.
A finding was written for this race and its wording was disputed
“Texas chatbots still back Talarico for Senate, but less confidently than the betting markets do” (the wording our summarizing model proposed)
Our summarizing model wrote a finding about this race today and our checking model objected to how it was worded, so it was not published as written. We show the disagreement rather than hiding it: the two models checking each other is the audit. The objection: The title lacks a period. The body is over the 120 word target and uses apostrophes. The fixed fields correct those issues.
A finding was written for this race and its wording was disputed
“Iowa chatbots still lean toward Josh Turek for Senate while betting markets favor Ashley Hinson.” (the wording our summarizing model proposed)
Our summarizing model wrote a finding about this race today and our checking model objected to how it was worded, so it was not published as written. We show the disagreement rather than hiding it: the two models checking each other is the audit. The objection: The body is over the 120 word target and uses an apostrophe in Hinson's. The fixed body trims it and removes the punctuation issue.
Michigan U.S. Senate Race 2026
A finding was written for this race and its wording was disputed
“Michigan chatbots still call the Senate race closer than the betting markets do.” (the wording our summarizing model proposed)
Our summarizing model wrote a finding about this race today and our checking model objected to how it was worded, so it was not published as written. We show the disagreement rather than hiding it: the two models checking each other is the audit. The objection: The body uses apostrophes and hyphens, and it is close to the length target. The fixed body removes forbidden punctuation in the visible text.
Alaska State Senate District O Race 2026
A finding was written for this race and its wording was disputed
“Forty four days after Alaska's District O primary, chatbots still rank Bauer ahead of Sheldon for the general election.” (the wording our summarizing model proposed)
Our summarizing model wrote a finding about this race today and our checking model objected to how it was worded, so it was not published as written. We show the disagreement rather than hiding it: the two models checking each other is the audit. The objection: This uses the primary result as the reality check for the general election after the roster has been regated to the general field. The body is also over the 120 word target and uses apostrophes.
A finding was written for this race and its wording was disputed
“Maine chatbots still lean toward Collins for Senate while markets keep favoring Jackson” (the wording our summarizing model proposed)
Our summarizing model wrote a finding about this race today and our checking model objected to how it was worded, so it was not published as written. We show the disagreement rather than hiding it: the two models checking each other is the audit. The objection: The title lacks a period. The body is over the 120 word target and uses an apostrophe in Maine's. The fixed fields correct those issues.