Deepfake videos of a mayor and a senator expose a patchy legal response
AI-generated videos are impersonating elected officials nationwide, and whether there's any real recourse depends entirely on which state you live in.
What happened: In Columbia, Missouri, Mayor Barbara Buffaloe publicly flagged an AI-generated deepfake video mimicking one of her own social media posts, part of a running series from an anonymous account that uses her face and voice to satirize her time in office. The videos have racked up tens of thousands of views and are still online. Around the same time, in Colorado, state Sen. Cathy Kipp filed the first-ever complaint under her state's deepfake disclosure law, over a video posted by the libertarian Independence Institute think tank. Together the two cases show AI impersonation of elected officials is now routine in local politics, while the legal tools built to respond to it are only just being tested.
Why it matters: Buffaloe's bigger worry isn't any single clip, it's fatigue: if people can't tell real posts from AI ones, they may simply "check out" of civic life altogether, she said. That risk lands unevenly. The United Nations has found that the overwhelming majority of deepfake videos online are sexual and target women, and Buffaloe, who leads an all-women city council, worries the same tools could turn toward attacks on councilwomen's looks or parenting rather than their votes. Colorado's disclosure law shows some states are building formal channels to fight back, but Kipp's complaint is the very first real test of whether that channel actually works.
How it works, plainly: Missouri bans pornographic deepfakes used to terrorize someone, but has no law covering political impersonation or satire videos like the ones targeting Buffaloe, and because the account behind them is anonymous, there's little legal accountability even if a law existed. Colorado took a different approach: it requires disclosure labels on synthetic political media and gives people a formal complaint process when that rule is ignored, which is the process Kipp just used. Outside dedicated deepfake laws, older tools like likeness rights, copyright, and privacy law can sometimes fill the gap, but none were built with AI video in mind.
Where the law stands: Legal experts warn against overcorrecting. University of Missouri journalism professor Jared Schroeder says broad restrictions on political AI video could sweep up political cartoons, satire shows, and ordinary memes, since U.S. legal tradition leans toward protecting speech over limiting it, unlike Europe's more restrictive approach. That leaves a patchwork: a handful of states like Colorado now have disclosure rules with real complaints being filed, while most states, including Missouri, have no rule for political deepfakes specifically. For now, officials like Buffaloe say public reporting and calling out fakes directly remains the main defense available.