The New Political Battleground: How Politicians Are Fighting to Control What AI Chatbots Say About Them

Political campaigns have something new to worry about these days: unflattering or incomplete information produced by A.I. programs when voters ask questions about candidates. Dustin Lloyd, a Democratic primary candidate for Missouri's state legislature, was a well-known member of his community when he started his political campaign. But there was one important group that didn't seem to know him at all: A.I. chatbots.

If voters quizzed a chatbot like OpenAI's ChatGPT or Google's Gemini about Lloyd, they would find only basic information that failed to convey his focus on helping small businesses. Fortunately for Lloyd, there was a fix. He tweaked his online presence, publishing a Q&A about himself on his campaign website. Later, when the chatbots were quizzed again, they dutifully connected that personal history to his policy goals. Editing A.I.'s output proved to be easier than he imagined. Since then, Lloyd said, "I'm constantly working on it every day."

Lloyd's story is not an isolated case. It represents a fundamental shift in how political campaigns must operate in the age of artificial intelligence. Candidates have long had to worry about their reputations among voters. Now, they have to worry about what A.I. thinks about them, too.

The New Industry: Managing Your AI Reputation

A new industry has sprung up to help candidates navigate this world. Lloyd relied on a report on his A.I. presence from CampSight, a tool launched by Run For Something Action Fund, a progressive group that recruits young candidates to run for office. It determined that Lloyd's website and his presence on Wikipedia and Ballotpedia sites where voters regularly find information about political figures were lackluster, advising him to make changes. It even recommended posting threads on Reddit so that the chatbots would begin including information from the online forum in their responses.

This represents a new frontier in political consulting. Campaigns are no longer just managing their message to voters through traditional media and advertising; they are now managing how algorithms interpret and present their candidates to the public. The stakes are high: millions of voters are turning to chatbots for election information. With 29.8% of Hungary's population identified as AI users, the researchers who tested AI tools during that country's election found that the responses often look confident and authoritative, making users more likely to trust them.

The Problem: When AI Gets It Wrong

The concern among politicians is not just that chatbots might present unflattering information, but that they might present false information. Research has consistently shown that AI chatbots provide inaccurate, inconsistent, and unreliable guidance to voters.

A study during Hungary's parliamentary elections found that AI chatbots recommended parties not running and gave highly volatile answers to identical prompts. In percentage-matching tests, ChatGPT assigned the eventual winning party a score in just 2% of cases. Instead, users with views aligned with that party were routinely recommended a smaller party unlikely to cross the parliamentary threshold, or parties not even on the national ballot. By contrast, the ruling party's profiles were recognised far more consistently.

Other research has found similar issues. ChatGPT omitted at least six candidates from by-election ballots when listing who was running in the constituencies. Google's Gemini incorrectly stated that Gerry Hutch had won a seat in the 2024 general election. A Demos study found that one third of responses across chatbots contained factual errors. Researchers also noted that both AI models regularly began their responses with polite disclaimers saying they "cannot give political advice" before providing several paragraphs of dense, highly persuasive party recommendations.

"The answers appeared well-argued, precise and authoritative," the researchers said. "This creates a risk that users may treat the outputs as reliable."

The Backlash: When Politicians Fight Back

Not all political uses of AI have been subtle reputation management. Some campaigns have taken a more aggressive approach with significant backlash.

California state Sen. Scott Wiener created an AI chatbot mimicking his opponent, San Francisco Supervisor Connie Chan, in the race for Rep. Nancy Pelosi's seat. "I'm Connie, the world's first AI trained to say 'no' to anything you ask," the chatbot site said. The bot appeared to mock Chan's record on issues like housing and public transit.

But the tactic backfired spectacularly. Pelosi and nine other female lawmakers condemned the strategy in a letter, asserting that it furthered "racist misogynistic tropes." "Women have spent generations fighting to have our own voices heard in the halls of power," the letter stated. "We should not have to fight against artificial intelligence putting words in our mouths."

The controversy centered on how the bot responded to questions about Chan's accent and citizenship status. Screenshots shared by Chan's campaign appeared to show the bot responding to user prompts about her accent with phrases including: "I'm afraid my San Francisco accent is so thick I can barely understand myself."

Wiener initially defended the chatbot, accusing Chan's campaign of "bombarding the chatbot with repeated malicious questions" and calling it "ironic" that Pelosi would express concern about AI when she had contributed to killing his first major piece of AI legislation. But he ultimately apologized and took down the chatbot. "Innovation means trying new things, and it also means being honest when something doesn't land the way you intended," he said.

The Brazil Example: AI Defying Election Rules

The challenge of controlling AI in elections extends beyond the United States. In Brazil, AI chatbots have continued answering political questions in defiance of new electoral rules banning them from giving voting tips. The country's electoral court has taken a leading role in the fight against disinformation, declaring far-right former president Jair Bolsonaro ineligible to run for office for spreading false information. But the chatbots have proven difficult to regulate.

Meanwhile, an AI-generated avatar of Bolsonaro who is under house arrest for plotting against democracy allowed him to make a surprise appearance at the launch of his son's presidential campaign. This raises profound questions about the ability of electoral authorities to control the use of AI in political campaigns, even when candidates themselves are barred from running.

The Trust Problem: Voters and AI

The stakes of these developments are significant. A March 2026 poll found that 85% of Americans say it is likely that AI-generated political content will spread misinformation about elections. Nearly one in three voters saw a deepfake or AI-generated video, audio, clip or image about an election candidate or politician online in the run-up to local council elections.

Yet voters often assume AI neutrality. They may not realize that chatbots can be manipulated, that they frequently hallucinate, or that their training data may contain biases. The researchers studying Hungary's election warned that "in a more competitive election or where voters are less certain, such outputs could potentially impact the outcome."

The 2026 election cycle represents a turning point. This will be the first election where AI is both a dominant policy debate and a widely deployed political weapon. AI allows both candidates and bad actors to persuade and perplex with unprecedented speed, scale, and precision.

For candidates like Dustin Lloyd, the solution is to learn how to influence what AI says about them. For others, like Scott Wiener, the experiment may backfire. For voters, the challenge is to recognize that when they ask a chatbot for political advice, they are not getting objective truth they are getting a product shaped by algorithms, training data, and increasingly, the strategic efforts of campaigns themselves.

As one researcher put it, "The results raise serious concerns about the reliability of general-purpose AI systems in electoral contexts." In a democracy, that is a concern that should trouble everyone.