Artificial intelligence has moved from a campaign talking point to a budget item. New federal spending disclosures show candidates, parties, and committees are paying AI companies for tools that now sit inside the daily work of American politics.
Itemized expenditure data from the US Federal Election Commission, covering spending dating back to 2020, shows at least $17m in disclosed payments to AI technology vendors across 523 federal candidates and campaigns. The figure does not capture every use of AI, but it offers a clear view of how campaigns are beginning to spend money on the technology.
ChatGPT and Claude Find Separate Political Audiences
OpenAI is one of the best-known names in the spending records. At least 80 federal campaigns and committees have reported spending with OpenAI since 2024, with total payments reaching about $50,000. The Republican National Committee is the largest overall buyer, with nearly $10,000 in reported expenses.
Other individual users include the campaigns of Republicans Mike Lawler, John Kennedy, and Bill Cassidy, along with California Democrats Ro Khanna and Ted Lieu. The names span both major parties, suggesting that AI tools are being adopted across the political landscape rather than by one ideological group.
Anthropic has seen a much newer wave of campaign interest. At least 65 candidates or committees reported paying the company in 2026, compared with essentially zero in previous years. Tom Cotton, a Republican senator from Arkansas, was the largest individual user, with his campaign reporting more than $4,000 in Anthropic software spending in June 2026.
Campaigns tend to choose either Claude or ChatGPT, rarely both. About 12% of campaigns or committees use both vendors, a pattern that points to separate tool preferences rather than a broad effort to collect every available AI service.
xAI also appears in the records, though on a smaller scale. Seven federal and two state-level candidates or committees have reported paying the company, with about $5,000 spent. Most of that amount came from RFK Jr’s 2024 presidential campaign.
Specialized Vendors Have Already Taken Bigger Budgets
The largest campaign AI payments do not come from the most famous general-purpose chatbot companies. AmplifAI provided $4.7m in campaign spending during the 2022 cycle, when Democratic candidates including Mark Kelly, Joe Biden, Bernie Sanders, and Adam Schiff used the company’s services.
Daisychain has drawn about $300,000 in reported candidate spending during the 2026 cycle. These figures show why the total AI market inside campaigns cannot be measured by chatbot subscriptions alone. Campaigns are also directing money toward vendors built for political work, alongside the tools used for everyday writing and analysis.
The spending arrives as lawmakers argue over whether the government should place new limits on AI. The debate has been pushed past the November midterms, leaving Congress with the prospect of another year without comprehensive AI legislation.
Some Democratic and Republican members say Congress needs to move faster on AI safety, national security, and the technology’s impact on critical infrastructure. GOP leaders have warned that rules could stifle innovation or weaken the United States in its competition with China.
Congress Faces a Narrow Window for AI Action
Speaker Mike Johnson has called for a “balanced approach” to regulation and warned that hyperregulation could cost the nation its edge. President Trump is calling for AI to be rebranded as “super intelligence” and has argued against establishing new guardrails.
Other lawmakers want tougher action. Sen. Bernie Sanders and Rep. Greg Casar introduced the Ban Artificial Superintelligence Act, which would ban the development or deployment of artificial superintelligence and create a federal Department of Artificial Intelligence.
Reps. Mike Lawler and Josh Gottheimer introduced the Stop Rogue AI Act, aimed at helping businesses and federal agencies spot and stop dangerous AI agents. Reps. Nathaniel Moran and Ted Lieu introduced the AI Kill Switch Act, which would require AI system developers to maintain the ability to throttle, suspend, or shut down their systems.
Sen. John Kennedy sought to pass legislation requiring tech companies to develop a “kill switch” for superintelligent AI, but Sen. Rand Paul blocked the request. Paul warned, “If Congress acts hastily before the technology is understood, Congress risks killing innovation and setting this technology back decades.”
Senate Majority Leader John Thune has called AI threats “real” and introduced an AI safety bill with Sen. Amy Klobuchar. Sanders has taken a stronger position, saying, “Scientists have been clear: AI is an existential threat to humanity. It must be treated as such.” He also said, “When the future of humanity is at stake, we cannot let a handful of Big Tech CEOs write their own rules.”
The timing leaves Congress with less than two months after the midterm elections to reach a deal before the new Congress is sworn in. Rep. Don Beyer said Democrats are expected to work in “good faith” with Republicans and Trump on AI regulation, adding, “We have a couple of weeks there, and we can actually do something meaningful.”
That political deadline now sits beside a practical reality: campaigns are already spending millions on AI tools. While lawmakers debate guardrails, candidates and committees are deciding which systems belong in their operations, how much those systems are worth, and whether they need one vendor or several.
Based on
