AI election headlines for Oct 10, 2026
What ChatGPT, Claude, Gemini and Grok told voters about the 2026 races on Oct 10, 2026, set against prediction markets and recent polls. Each finding was written by one model and checked by another.
Alaska betting markets edge toward Sullivan, but chatbots and bettors still split on the Senate race
Chatbots picked Republican incumbent Dan Sullivan 55.2 percent of the time when Alaska voters asked about the U.S. Senate race, while live prediction markets give Democrat Mary Peltola a 73.5 percent chance to win. Markets shifted 3.5 points toward Sullivan since yesterday but still give him only a 27.1 percent chance. Recent polls are closer, putting Sullivan at 44.6 percent of the vote and Peltola at 43.5 percent, 24 days before the general election.
Alaska chatbots still pick Kreiss-Tomkins for governor, while markets are more confident
Asked as Alaska voters, chatbots picked Democrat Jonathan Kreiss-Tomkins 54.3 percent of the time. Markets give him a 77.5 percent chance to win, while polls put him at 31.7 percent of the vote. None of the three measures moved much since Friday, so the useful signal is the same steady split voters have seen all week. The chatbots favor the same front runner as bettors, but with less certainty, and the latest poll average is lower still.
Alaska U.S. House At-Large Race 2026
Alaska chatbots still pick Begich for House, with markets more confident and polls closer
Chatbots picked incumbent Nick Begich III (R) 65.5 percent of the time when Alaska voters asked about the U.S. House At-Large race. Markets give Begich an 86.0 percent chance to win, while polls put him at 47.2 percent of the vote. Bill Hill, an Independent, is the main reason the lenses differ. Chatbots picked Hill 14.7 percent of the time, but polls put him at 37.4 percent. With no meaningful move since yesterday and the same Begich edge reported repeatedly this week, the race looks steady 24 days before the general election.
Iowa chatbots still give Josh Turek a small Senate edge, while markets favor Ashley Hinson
Iowa voters asking chatbots about the U.S. Senate race would still see Democrat Josh Turek picked slightly more often than Republican Ashley Hinson. Chatbots picked Turek 44.6 percent of the time and Hinson 38.8 percent. Betting markets give Hinson a 56.5 percent chance to win, while markets give Turek 43.5 percent. Polls put Turek at 46.5 percent of the vote and Hinson at 43.9 percent, with the latest poll 17 days old. None of the three measures changed much since Friday, so this is the same split Iowa voters have been seeing for days.
Ohio U.S. Senate Special Election 2026
Ohio chatbots still pick Sherrod Brown for the special Senate race, with the race largely steady
Chatbots picked Democrat Sherrod Brown 52.4 percent of the time when Ohio voters asked about the U.S. Senate special election. Markets give Brown a 59.5 percent chance to win, and polls put him at 47.8 percent of the vote. Incumbent Jon Husted (R) remains the main mismatch, with chatbots picking him 30.0 percent of the time while polls put him at 42.9 percent. None of the three measures moved enough since Friday to make this a new turn in the race, 24 days before the general election.
Texas chatbots still pick James Talarico for Senate, but betting markets are more confident
Chatbots picked Democrat James Talarico 46.4 percent of the time when Texas voters asked about the U.S. Senate race. Betting markets give him a 62.5 percent chance to win, a 16.1 point gap. Polls put Talarico at 48.6 percent of the vote, closer to the chatbot read, with Republican Ken Paxton at 44.3 percent. Nothing moved much since yesterday, so this is a steady race picture rather than a new break.
Michigan U.S. Senate Race 2026
Michigan chatbots still make the Senate race look closer than betting markets do
Michigan voters asking chatbots about the U.S. Senate race are getting a tighter picture than the betting markets show. Chatbots picked Democrat Abdul El-Sayed 50.4 percent of the time and Republican Mike Rogers 47.4 percent. Markets give El-Sayed a 66.5 percent chance to win and Rogers a 33.5 percent chance. Polls put El-Sayed at 46.9 percent of the vote and Rogers at 43.7 percent. Nothing moved much since Friday, so this is the same steady split Michigan voters have been seeing all week.
Maine chatbots are nearly split on the Senate race, and the other measures also show a close contest
Twenty four days before the general election, chatbots picked Democrat Troy Jackson 50.2 percent of the time and incumbent Susan Collins (R) 49.6 percent. Markets give Jackson a 57.3 percent chance to win, while polls put Collins at 47.5 percent of the vote and Jackson at 46.5 percent. Nothing moved much since Friday, so the useful signal today is steadiness. A Maine voter asking a chatbot about the Senate race would likely see a near tossup, not a clear recommendation.
Alaska State Senate District O Race 2026
District O chatbots still favor incumbent George Rauscher with no live market or poll to check them
Alaska voters asking chatbots about State Senate District O are still most likely to be pointed to incumbent George Rauscher (R). Chatbots picked Rauscher 48.2 percent of the time, Democrat Peter Bauer 35.4 percent and Republican Ryan Sheldon 14.3 percent. There is no live prediction market or polling in the data for this race, so the chatbot panel is the only current measure here. The August 18 top four primary advanced all three candidates to the general election still to come. In that primary, voters put Rauscher first at 38.8 percent, then Sheldon and Bauer tied at 30.6 percent each.
Alaska State Senate District O Race 2026
A finding was written for this race and its wording was disputed
“District O chatbots still pick incumbent George Rauscher with no live market or poll to check them” (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 AK-SS-O rank agreement data shows verdict partial with an inversion: chatbots rank Bauer above Sheldon (35.4 vs 14.3) but the primary had Sheldon and Bauer tied at 30.6 each, and the computed biggest_level_miss flags Sheldon being badly underrated by the panel (14.3 AI vs 30.6 actual, a 16.3pp gap). The body omits this inversion entirely, presenting the chatbot order as if uncontested. Per the rank-agreement override rule, a partial finding must name the inversion. The original phrasing 'all three advanced' followed by 'is history now' is acceptable framing, but the missing inversion disclosure is the dispute. Also 'pick' used as present tense verb for an ongoing panel read is fine, but the fix retitles to 'favor' only for clarity, not a rule violation.