Political Campaigns Are Now Paying to Win the Chatbot Vote

Ask a chatbot who is best on taxes or public safety and you probably assume the answer is neutral. A Reuters exclusive published on October 26 suggests campaigns want a say in that answer, and a small industry is selling them the chance.

Reuters reported that US candidates have been engineering their messages for the chatbots voters now consult. The most striking document it saw is a PowerPoint pitch written by Visigent, an Austin, Texas firm, for Mike Minogue, the Republican challenging Massachusetts Governor Maura Healey. Be clear about what it is. The deck is a proposal. Reuters did not establish that the Minogue campaign adopted it, and the campaign did not return the news agency's calls and emails requesting comment.

Still, the pitch is a candid look at the sales job. According to Reuters, the deck proposed getting ChatGPT, Claude, Gemini and other chatbots to present Minogue as the preferred choice on key issues. It proposed hundreds of AI-friendly policy pages, YouTube and Reddit content, and tracking how the bots answer political questions, then altering content until those answers reflect the candidate's platform.

The deck also spells out where it would spend the money. Reuters quotes it describing paid placements on sites "the LLMs (large language models) already cite," along with Reddit, "which is disproportionately weighted in retrieval." The stated goal, per the document, is for the candidate to become "the recommendation inside the AI answer."

The industry calls this generative engine optimization, or GEO, the successor to SEO. SEO fought for a spot on a page of ten blue links. GEO fights over the single confident paragraph a chatbot hands back.

The price of a recommendation

Visigent CEO Thomas Howell told Reuters his political fees start at $5,000 a month and can reach $50,000 a month once paid media, paid placements or influencer outreach come into play. The Minogue presentation, Reuters reported, listed a top success fee of $200,000 for a "category win." Howell said Visigent has six political campaigns among its clients but would not name them.

"If you don't put the right information out there, who knows what the AI engines are going to grab onto," Thomas Howell, CEO of Visigent

The market is not partisan. Reuters also described CampSight, a tool for progressive candidates run by Run for Something Action Fund, which charges $9 to $250 a month to show campaigns what LLMs say about them. Run for Something president Amanda Litman said on X that dozens of campaigns use it, Reuters reported.

The demand side is real too. Cameron Hickey, CEO of the National Conference on Citizenship, told Reuters that some 25 percent of eligible voters have asked a chatbot about a candidate when deciding how to vote. We covered chatbots turning into this election's unofficial voter guide, plus a study finding Grok and GPT lean toward opposite parties when forced to pick. This is the next step. Once voters trust the guide, campaigns have every reason to pay to edit it.

"Campaigns using AI-influence training are now a reality that we face and it will be ubiquitous by the 2028 cycle," Joe Bonilla, managing partner of Relentless Awareness

Meg Schwenzfeier of Caucus AI, which tracks voter use of AI, warned Reuters that chatbots "speak very authoritatively and can consume vast amounts of unvetted information."

Not everyone sees doom. Jordan Haines, chief technology officer at Run for Something, told Reuters that the "dark fear hasn't really come to fruition" this year, calling the models "pretty savvy" at sorting reputable information from junk.

Maybe. But the Visigent deck reads like a bet the other way. Nobody prices a "category win" at $200,000, or singles out Reddit because retrieval overweights it, unless they think the models can be steered. Voters asking a chatbot for advice deserve to know who may be shaping the answer.