There is a revolution underway in American politics. It is not happening in the marble corridors of Congress. It is not being debated on the Senate floor. It is happening in campaign war rooms and consulting firms from Boise to Birmingham, where artificial intelligence has moved from curiosity to cornerstone. And Washington, true to form, is responding with the one tool it knows best: the blunt, clumsy instrument of regulation that manages simultaneously to overreach and to miss the point.
The numbers are not subtle. A new survey from the American Association of Political Consultants tells us that 83 percent of political professionals now use AI tools at least once a week. Last year, that figure was 59 percent. Nearly half of all consultants, 48 percent, are using these tools multiple times a day. That is not a trend. That is a transformation. And it is happening faster than any government body has the wit or the will to understand. 1
Republicans, it should be noted, are leading this charge. Nearly two-thirds of GOP consultants report using AI daily. Their Democratic counterparts clock in at 49 percent. Republicans are also more likely to call the technology genuinely effective, at 40 percent, compared with Democrats’ 31 percent. This is not a coincidence. The Republican Party has, for better and worse, always been more willing to trust markets and technology over mandarins and memoranda. In this race, that instinct is paying dividends.
“The blunt instrument of regulation manages to overreach and miss the point at the same time.”
The Disclaimer Debacle
But here is where the story gets genuinely troubling. Because while consultants are sprinting forward, state legislatures across the country have been busy constructing a legal obstacle course of AI disclosure requirements. The intention sounds noble. Be transparent. Let the voter know when artificial intelligence had a hand in producing a political advertisement. Who could argue with that?
You could argue with it, and you should, after reading what the AAPC Foundation found when they studied the actual effect of these disclaimers on real voters. The conclusion is damning. When an AI disclosure label appears on a political ad, viewer approval of the ad’s message declines sharply. Trust craters. Skepticism spikes. And here is the kicker: this happens whether the ad contains any AI-generated content or not.2 The disclaimer itself poisons the well. It does not inform the voter. It panics him.
The researchers called it a cognitive speed bump. That is a polite academic way of saying that well-meaning legislators have handed every political opponent a weapon. Slap an AI disclaimer on an ad, and suddenly the audience tunes in with crossed arms and narrowed eyes. The message is already half-dead before the candidate says a word. This is not transparency. This is regulatory sabotage dressed up as good governance.
AAPC’s Julie Sweet put it plainly. The language policymakers are putting into law has a significant effect, as do the formatting requirements. Bigger type means more distrust. Smaller type goes unnoticed. So the disclosure accomplishes nothing except harm when it is done right and nothing at all when it is done poorly. This is government competence in the year 2026.
The Patchwork Problem
The compliance landscape is, to use the kindest possible word, chaotic. At least 26 states have now enacted laws governing AI in political advertising.3 Each one is slightly different. Each one creates its own liability regime, its own disclosure standard, its own definitions. A campaign running a statewide race in one state faces an entirely different legal universe than the same campaign operating in another state. Maine proposes fines up to five times the cost of an ad buy. That is not a deterrent. That is a financial execution for a campaign that makes an honest mistake.4
Meanwhile, at the federal level, the picture is no clearer. The FCC has proposed its own disclosure requirements for radio and television.5 The FEC has a petition for rulemaking sitting in its inbox that it has not resolved. Congress introduced the REAL Political Advertisements Act and then let it die without a vote.6 So we have 26 different state regimes, a pending FCC proposal, a stalled FEC process, and a dead bill. That is not a regulatory framework. That is a bureaucratic fog bank.
The burden falls hardest not on the well-funded national campaigns with platoons of lawyers. It falls on the state legislative candidate, the county commissioner hopeful, the first-time challenger running on a shoestring who cannot afford to hire a compliance team to navigate this patchwork. In other words, the regulations built in the name of protecting democracy are doing the most damage to the democratic process at its most local and human level. This is how Washington works. This is what it produces.
“This is not transparency. This is regulatory sabotage dressed in the language of good governance.”
The Case for Self-Regulation
Republicans understand this instinctively, which is why 57 percent of GOP consultants believe AI use should be self-regulated rather than governed by Washington.7 They are not wrong. The technology is moving faster than any legislature can track. By the time a bill is drafted, debated, amended, passed, signed, and promulgated, the AI tools it was written to govern will have gone through three or four generations of development. Regulation that is perpetually behind the curve is not protection. It is theater.
The better approach is the one the marketplace is already developing. Platforms like Google have implemented their own AI disclosure requirements.8 Industry bodies like the AAPC are producing research and guidelines. Consulting firms are building internal standards. These mechanisms are imperfect. They will make mistakes. But they are adaptive in ways that statute never can be. They respond to evidence. Legislation responds to lobbying and election cycles.
None of this is an argument for lawlessness. Nobody serious is suggesting that deepfakes ought to be consequence-free or that AI-generated audio mimicking a rival politician is simply vigorous campaigning. The FCC was right to fine the consultant who used an AI voice to impersonate a president.9 There are legitimate cases for law. The question is whether panicked, patchwork, one-size-fits-all mandates represent those cases. They do not.
What This Means for the Republic
Zoom out for a moment and consider what we are watching. American political campaigns are in the midst of perhaps the most dramatic technological transformation since the invention of television advertising. The tools available to a consultant in 2026 would have seemed like science fiction in 2016. AI is writing messaging, targeting audiences, analyzing polling data, and producing creative content at a cost and speed that simply has no precedent. The democratizing potential is enormous. The risks are real. And the response from the political class is to slap a warning label on it and call it a day.
Forty-three percent of political consultants now believe AI will have a positive impact on the future of campaigns. That is up from 33 percent a year ago.10 The industry is becoming more optimistic even as the regulatory environment grows more hostile. That gap tells you something important about where real knowledge of this technology actually resides. It does not reside in state legislatures. It resides in the firms doing the work.
The voter, caught in the middle, is left to navigate a world where the tools are advancing rapidly, the rules are a mess, and the warnings on the screen may mean everything or nothing depending on which state he lives in and how big the font is. That is not good enough. The American voter deserves better. He deserves a regulatory conversation that is serious, evidence-based, and honest about the limits of what government can actually accomplish in a domain that changes faster than the law can move.
Conclusion
Here is the bottom line. AI is not going away. It is not slowing down. It is embedding itself into every corner of American political life, from the ad you see on your phone to the fundraising email in your inbox to the opposition research memo sitting on a campaign manager’s desk. The only question is whether the rules we build around it will be smart or stupid, adaptive or rigid, honest or performative.
Republicans, to their credit, seem to understand that the market and the industry are better positioned to answer that question than the federal government. Democrats, to their detriment, keep reaching for the same lever they always reach for: more rules, more mandates, more bureaucracy, more liability. And the evidence, as the AAPC Foundation has now documented, is that the lever they are pulling is making things worse.
That should matter to everyone, regardless of party. A democracy that cannot have an honest conversation about how its campaigns are run is a democracy in trouble. A regulatory regime that penalizes transparency without delivering it is worse than no regime at all. America built the greatest republic in human history on the principle that free people, given good information, will generally make sound decisions. Give them bad information wrapped in a government mandate, and you get something quite different. We should all want better than that.
ENDNOTES
1. AI Use Among Political Consultants is Booming. Here’s How, Campaigns & Elections. https://campaignsandelections.com/industry-news/ai-use-among-political-consultants-is-booming/
2. AI Disclaimers on Campaign Ads Are Sowing Mistrust, Study Finds, Campaigns & Elections. https://campaignsandelections.com/industry-news/ai-disclaimers-on-campaign-ads-are-sowing-mistrust/
3. AI Ad Disclosure Laws Create New Compliance Minefield for Campaigns, Campaign Now. https://www.campaignnow.com/blog/ai-ad-disclosure-laws-create-new-compliance-minefield-for-campaigns
4. AI Ad Disclosure Laws Create New Compliance Minefield for Campaigns, Campaign Now. https://www.campaignnow.com/blog/ai-ad-disclosure-laws-create-new-compliance-minefield-for-campaigns
5. FCC Proposes Disclosure Rules for the Use of AI in Political Ads, Federal Communications Commission. https://www.fcc.gov/document/fcc-proposes-disclosure-rules-use-ai-political-ads
6. AI-Generated Ads Are Trickling Into Political Campaigns, NBC News. https://www.nbcnews.com/politics/2028-election/ai-midterms-politics-campaign-ads-rcna263752
7. AI Use Among Political Consultants is Booming. Here’s How, Campaigns & Elections. https://campaignsandelections.com/industry-news/ai-use-among-political-consultants-is-booming/
8. The Ultimate Guide to Political Advertising in 2026, Basis Technologies. https://basis.com/blog/the-ultimate-guide-to-political-advertising-in-2026
9. Real or Fake? AI is Blurring the Line in Political Ads, KHOU News. https://www.khou.com/article/news/politics/real-fake-ai-political-ads/285-cbdd5a03-e9bd-4a4c-918e-35f89322ab4e
10. AI Use Among Political Consultants is Booming. Here’s How, Campaigns & Elections. https://campaignsandelections.com/industry-news/ai-use-among-political-consultants-is-booming/
Dwight Patel is a contributing columnist for Direct Line News covering political technology, campaign strategy, media policy and Vice Chairman of the MDGOP