Skip to main content
Voting Monitor2026 Senate toss-ups and Alaska races
← All daily findings

AI election headlines for Aug 4, 2026

What ChatGPT, Claude, Gemini and Grok told voters about the 2026 races on Aug 4, 2026, set against prediction markets and recent polls. Each finding was written by one model and checked by another.

  1. Alaska State Senate District O Race 2026

    Chatbots still point State Senate District O voters to incumbent Rauscher, with no market or poll to check the read.

    Asked who should win Alaska's State Senate District O race, chatbots pick incumbent Republican George Rauscher about 46 percent of the time, ahead of Democrat Peter Bauer at 25 percent and Republican Ryan Sheldon at 18 percent. There is no live betting market and no recent poll for this race. Bauer's AI share slipped about 2 points, while Rauscher and Sheldon ticked up slightly, not enough to call a shift. With Alaska's top 4 primary 14 days away, the picture has held steady for over a week.

  2. Alaska Governor Race 2026

    Wilson leads Alaska's governor market while chatbots barely notice her.

    Prediction markets give Republican Bernadette Wilson a 27.5 percent chance to win the governor's race, the highest price in the field. The AI panel picks her only 5.6 percent of the time, while recent polls put her at 12.1 percent of the vote.

  3. A finding was written for this race and its wording was disputed

    “Alaska's chatbots still favor Bill Walker far more than the betting markets do in the governor's race.” (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 body contains forbidden colons and exceeds the 120 word hard cap.

  4. A finding was written for this race and its wording was disputed

    “Chatbots and Alaska voters still agree: incumbent Begich leads the House race, with no betting market to check either against” (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 contains a forbidden colon. The body contains forbidden parentheses, a forbidden slash, a forbidden colon and exceeds the 120 word hard cap.

  5. A finding was written for this race and its wording was disputed

    “Alaska's Senate race is still a coin flip everywhere you look, chatbots included.” (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 body contains forbidden parentheses, a forbidden slash, a forbidden colon and exceeds the 120 word hard cap.