On a recent afternoon in York, Pennsylvania, Alex Bond listed the political issues weighing most heavily on his mind. After gas prices and taxes, he cited data centers — the sprawling facilities that consume vast amounts of energy to power artificial intelligence.
“A.I. is terrible,” said Mr. Bond, a 29-year-old account manager at a firm supplying ankle monitors. “And it’s probably going to kill us all.”
He was sharing these views with a pair of canvassers deployed to Pennsylvania’s competitive 10th Congressional District by the Democratic nominee, Janelle Stelson, as part of a traditional door-knocking effort organized by the progressive group Swing Left.
What Mr. Bond did not realize was that his comments would soon be processed inside one of those very AI data centers.
After the conversation ended, one of the canvassers dictated a summary into a phone app. That memo was then analyzed by an AI tool alongside hundreds of others collected from similar interactions across the district. The software synthesized voter sentiment, generating reports the campaign could use to tailor messaging and convert persuadable voters like Mr. Bond into supporters.
“Everything a person is saying is a data point,” said Violet Kopp, a canvasser and Swing Left’s organizing program manager for the East Coast.
Republican and Democratic operatives who were once wary of artificial intelligence are now racing to integrate AI into their campaigns. Those who succeed could gain a decisive edge in the November midterms and, ultimately, the 2028 presidential race. Those who hesitate risk falling irretrievably behind.
Yet this seismic shift is colliding with significant resistance. Voters and campaign staff alike express deep distrust of the technology, fearing its potential to eliminate jobs, devour energy resources, and exacerbate climate change.
Polling indicates Democrats are more skeptical of AI tools than Republicans. Progressive strategists have struggled to deploy the technology without alarming volunteers or unionized staff concerned about job displacement. Republican strategists report fewer internal complaints, though conservative voters remain broadly apprehensive about AI’s societal impact.
Candidates find themselves caught in the middle, eager to harness AI’s efficiency while avoiding the political fallout of its perceived costs.
“It’s a political liability,” said Eric Wilson, a Republican strategist and director of the Center for Campaign Innovation, a nonprofit dedicated to modernizing conservative campaigns. “If voters don’t like A.I., they don’t want to know that their candidate’s campaign is using A.I. to draft emails, create press releases, or edit videos. So you’re just not going to see people bragging about it. But it is happening.”
AI-generated videos and images represent the public face of this overhaul. Spencer Pratt, a Republican who ran in this month’s Los Angeles mayoral primary, used AI content to mock opponents and conjure dystopian visions of the city. He ultimately lost the race.
Behind the scenes, however, campaign managers have rapidly embedded AI into nearly every operational tool, using it to analyze voter data, produce campaign materials, and write customized messages for micro-targeted segments of the electorate.
A new survey from the technology and politics newsletter Anchor Change found that 87 percent of campaigners and strategists now use AI daily. The tools dominate monthly conference calls hosted by a progressive nonprofit, drawing hundreds of participants who exchange tactics for improving operations and drafting targeted outreach.
Opposition researchers are likewise leveraging AI to comb through campaign finance records and surface damaging narratives. American Bridge 21st Century, a Democratic-aligned super PAC, has used AI to investigate roughly 250 Republican candidates and officeholders. The group credits the technology with enabling deeper dives into 17 House Republicans sitting in relatively safe districts. “A.I. is surfacing flags to humans, who vet the projects, expand on them, and get them out the door,” said Pat Dennis, the organization’s president.
AI adoption is a top priority for Kate Gage, executive director of the Higher Ground Institute, a progressive incubator pushing organizations to embrace new technology.
“Basically my whole life right now is figuring out how to get campaigns to identify what the use cases are,” she said.
Ms. Gage’s group hosts monthly digital “A.I. open mics” attended by hundreds of strategists who share methods for modernizing traditional campaign workflows. That surge of interest has populated an online database cataloging nearly 100 distinct AI tools for drafting messages and managing voter data. One documented use case describes converting a policy brief into a week’s worth of social media posts using ChatGPT and similar platforms.
Ms. Gage said the tools have revolutionized political organizing, granting campaigns the ability to conduct sophisticated research at unprecedented speed. “We’ve had a lot of unstructured information and data, but we haven’t been able to analyze it.”
Her work is part of a broader effort to persuade candidates, strategists, and party officials to adopt AI in ways that fundamentally transform campaigns. Advocates view the midterms as the ideal testing ground to build the infrastructure that will power the next presidential election.
The primary campaign for Saikat Chakrabarti, a Democratic candidate in California’s 11th Congressional District, replaced nearly every standard paid tool — from canvassing apps to phone-banking software — with proprietary versions built by just three staffers using AI.
The transition has not been seamless. One progressive organization reported receiving angry emails from supporters upset over AI’s intrusion into campaign operations.
Such friction could slow Democratic adoption while Republican campaigns accelerate, potentially handing the GOP an advantage in razor-thin contests.
“I think it’s a strategic error on their part,” Mr. Wilson said. “Whereas Republicans are saying, ‘OK, how can we use this to help us win and give us an advantage?’”
Like Ms. Gage, Mr. Wilson trains political operatives on AI implementation.
The Republican Party has moved aggressively to embed AI across its operations, Mr. Wilson noted, relying heavily on well-funded private vendors rather than the nonprofit-driven model favored by Democrats.
The two parties have also diverged on ethical guidelines. In a blog post on AI ethics, Mr. Wilson argued that generating AI videos of an opponent is acceptable provided they accurately reflect the candidate’s actual statements. By contrast, a training course from the National Democratic Training Committee instructs campaigners never to create such content, warning that it “undermines democratic discourse and voter trust.”