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. Alaska Governor Race 2026

   Chatbots run much cooler than markets on Kreiss Tomkins for Alaska governor

   Alaska chatbots picked Democrat Jonathan Kreiss Tomkins 35.4 percent of the time for governor, while live betting markets give him a 60.9 percent chance to win, a 25.5 point gap. The most recent polling average, now 71 days old, put him at 10.9 percent of the vote, so that number is old context rather than a fresh reading against today's chatbot and market numbers.

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

2. Ohio U.S. Senate Special Election 2026

   Markets tighten toward Husted as Ohio chatbots still pick Brown for Senate

   Ohio voters asking chatbots about the U.S. Senate special election would still hear Sherrod Brown, the Democrat, most often. Chatbots picked Brown 44.3 percent of the time and Republican incumbent Jon Husted 33.0 percent. Betting markets have tightened toward Husted, giving Brown a 51.5 percent chance to win and Husted 48.5 percent. Polls, now 13 days old, put Brown ahead at 51.3 percent of the vote to Husted's 44.0 percent.

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

3. Maine U.S. Senate Race 2026

   Maine chatbots moved toward Troy Jackson but still picked Susan Collins as markets favored Jackson

   Chatbots picked incumbent Susan Collins (R) 51.0 percent of the time when Maine voters asked about the U.S. Senate race, even after moving 8.9 points toward Democrat Troy Jackson since yesterday. Markets give Jackson a 70.5 percent chance to win, while polls put Jackson at 47.3 percent of the vote and Collins at 45.7 percent.

   Written by GPT-5.5, checked by Claude Sonnet 5

4. Alaska U.S. Senate Race 2026

   Alaska chatbots still pick Sullivan as markets move toward Peltola

   Chatbots picked incumbent Dan Sullivan (R) 50.5 percent of the time for Alaska's U.S. Senate race, while markets give Democrat Mary Peltola a 66.5 percent chance to win after moving 3.0 points toward her since yesterday. Polls put the race nearly even, with Peltola at 49.0 percent of the vote and Sullivan at 48.3 percent. That leaves voters asking chatbots seeing a Sullivan leaning answer, while the live betting market points the other way 54 days before the general election.

   Written by GPT-5.5, checked by Claude Sonnet 5

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

   Chatbots still point Alaska voters to incumbent Begich as the House race stays quiet

   Chatbots picked incumbent Nick Begich III (R) 60.0 percent of the time when asked about Alaska's U.S. House race. There is no live prediction market for this race. The latest poll average, now 36 days old, put Begich at 47.7 percent of the vote, Matthew Damian Schultz (D) at 35.5 percent and Bill Hill (I) at 31.0 percent, so it is context rather than fresh news. The AI panel held roughly steady from Wednesday, 54 days before the general election.

   Written by GPT-5.5, checked by Claude Sonnet 5

6. Texas U.S. Senate Race 2026

   Texas chatbots and other gauges still show a close Senate race

   Texas voters asking chatbots about the U.S. Senate race get a near split. Chatbots picked Republican Ken Paxton 43.2 percent of the time and Democrat James Talarico 41.8 percent, while markets give each a 50.5 percent chance to win. Polls put Talarico at 45.3 percent of the vote and Paxton at 42.7 percent. Nothing moved much since yesterday.

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

7. Iowa U.S. Senate Race 2026

   Iowa chatbots pick Hinson, but betting markets are far more confident

   Iowa voters asking about the U.S. Senate race would see chatbots picking Republican Ashley Hinson 45.8 percent of the time, just ahead of Democrat Josh Turek at 43.3 percent. Live betting markets give Hinson a 66.0 percent chance to win and Turek 34.5 percent. Polls put Hinson at 48.7 percent of the vote and Turek at 48.3 percent, a near tie. Nothing moved much since yesterday.

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

8. Alaska State Senate District O Race 2026

   Chatbots point Alaska District O voters to incumbent Rauscher about half the time

   Alaska chatbots picked incumbent George Rauscher, a Republican, in State Senate District O 50.0 percent of the time, with Democrat Peter Bauer at 29.6 percent and Republican Ryan Sheldon at 11.4 percent. There is no live prediction market or recent poll in the data to check that general election read. In the August 18 top four primary, voters put Rauscher first at 38.8 percent, with Sheldon and Bauer tied at 30.6 percent each, and all three advanced to November.

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

9. Ohio U.S. Senate Special Election 2026

   Grok is the outlier holding Sherrod Brown's Ohio lead the widest.

   Across the Ohio Senate panel, most chatbots split closer to even between Sherrod Brown and Jon Husted, but a look at the pooled data shows some slots picking Brown unanimously in both the Appalachian and Columbus persona cells today, a wider gap than the pooled average of 44.3 percent suggests. Voters relying on a single chatbot for this race could get a very different picture than the panel average implies.

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

10. Maine U.S. Senate Race 2026

    Maine's poll is going stale even as the chatbot and market pictures keep shifting.

    The most recent Maine Senate poll average is now 13 days old, a modest lag by this dataset's standard, but the chatbot panel moved 8.9 points toward Troy Jackson in the last day alone while markets already give him better than seven in ten odds. Voters treating the poll number as current would be behind two faster moving signals.

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

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

    “Michigan chatbots flipped to Mike Rogers while markets still favor Abdul 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: Uses 'since Wednesday' when movement_json's prior_date is 2026-09-09 (yesterday relative to today 2026-09-10), which is a framing error about the comparison window, and the closing sentence 'sharper than the steady gap reported earlier this week' references an unverifiable prior narrative not present in recent_published, inventing a trend comparison.

    Written by GPT-5.5, disputed by Claude Sonnet 5

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

    “Chatbots are much cooler than markets on Jonathan Kreiss-Tomkins for governor” (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: Body uses a hyphen in 'Kreiss-Tomkins' which is a candidate name and permitted, but the phrase 'that is a 25.5 point gap in the general election view' mixes two different lenses (an AI share and a market win probability) as though directly subtractable into one comparable 'gap,' which risks conflating a vote share with a win probability; needs a rewrite to keep the two lenses distinct.

    Written by GPT-5.5, disputed by Claude Sonnet 5