Election Deepfake Laws: What Is Banned, What Needs a Label, and Where

Two days before the 2024 New Hampshire primary, voters got a robocall in what sounded like President Biden's voice, telling them not to vote. It was an AI clone. The Federal Communications Commission proposed a $6 million fine, and the moment political deepfakes stopped being a hypothetical and became a legal problem with real penalties. This guide explains election deepfake laws in plain English: what is banned, what just needs a label, and where.

Disclaimer: This article is general legal information, not legal advice. Election deepfake laws change quickly, state counts shift every legislative session, and courts are actively reshaping what is enforceable. How any rule applies depends on your facts and your state. If you are dealing with a real situation, consult a licensed attorney.

Two days before the 2024 New Hampshire primary, voters got a robocall in what sounded like President Biden's voice, telling them not to vote. It was an AI clone. The Federal Communications Commission proposed a $6 million fine, and the moment political deepfakes stopped being a hypothetical and became a legal problem with real penalties. This guide explains election deepfake laws in plain English: what is banned, what just needs a label, and where.

Quick answer: There is no blanket federal ban on election deepfakes, but the FCC has ruled that AI-voice robocalls are illegal, and 30 states now regulate deepfakes in campaigns. Most state laws either require clear disclosure on AI-altered political ads or restrict deceptive candidate deepfakes close to an election, with exceptions for satire and news.

State counts in this area change every legislative session, and several laws are tied up in court, so treat every number below as a snapshot reviewed June 2026.

Federal Election Deepfake Rules

There is no single federal statute that bans election deepfakes. Instead, the federal picture is built from one firm rule on robocalls, an unresolved debate at the elections agency, and several pending bills.

The FCC AI Robocall Ruling

The clearest federal rule comes from the FCC. In February 2024, the Commission issued a declaratory ruling that AI-generated voices in robocalls are "artificial" under the Telephone Consumer Protection Act, which makes voice-cloned robocalls illegal without the called party's prior consent, per the FCC's announcement. The ruling took effect immediately and gave state attorneys general a direct tool against AI-voice calls.

The New Hampshire case showed the rule has teeth. The FCC proposed a $6 million fine against the political consultant who instigated the calls, and reached a separate $1 million settlement with the telecom carrier that transmitted them, per the FCC enforcement documents. Note that this rule governs how a deepfake is delivered (an automated phone call), not political content in general.

The FEC and AI in Campaign Ads

The Federal Election Commission, which regulates campaign finance and advertising, has not adopted a deepfake-specific rule. The FEC considered a petition asking it to clarify that its existing "fraudulent misrepresentation" rules cover deliberately deceptive AI in candidate ads. On September 26, 2024, the FEC declined to launch a deepfake-specific rulemaking, taking the position that its existing fraudulent-misrepresentation authority is sufficient. See the FEC's official site.

Pending Federal Legislation

Several bills addressing election deepfakes have been introduced in Congress, including proposals to require disclaimers on AI-generated political ads and to restrict deceptive synthetic media about federal candidates. None had been enacted into law as of June 2026. Pending bills can be tracked on congress.gov. Do not assume any specific bill will pass; treat all of them as pending until signed.

State Election Deepfake Laws: Three Approaches

States moved faster than Congress. As of May 2026, 30 states have enacted a law addressing deceptive synthetic media in elections, with Maryland becoming the 30th, per Public Citizen's election-deepfakes tracker. The laws fall into three broad approaches. For the state-by-state breakdown, see our deepfake laws by state table.

Ban States: Deceptive Candidate Deepfakes Prohibited

Some states prohibit the distribution of materially deceptive synthetic media of a candidate, usually within a defined window before an election and usually limited to content created with intent to harm a candidate or deceive voters. These are not blanket bans on all political AI; they target deception, and they typically carve out satire, parody, and news.

Disclosure States: Label It and It Is Legal

The most common approach is disclosure rather than prohibition. These laws let campaigns use AI-generated or AI-altered content as long as it carries a clear disclaimer telling viewers the media was synthetically created or manipulated. Label it accurately and it is generally lawful. Most states with a political deepfake law use this disclosure model.

Window States: Restrictions Near Election Day

Many statutes, whether ban-style or disclosure-style, only apply within a set window before an election, commonly 60 or 90 days. Outside that window, the restriction may not apply at all. The window reflects a legislative judgment that deception is most damaging when voters have no time to discover the truth before casting a ballot.

What Counts as an Illegal Election Deepfake

Across the state laws, the same elements tend to recur. A political deepfake is most likely to be unlawful when it combines a deceptive depiction, harmful intent, and bad timing, and is not covered by an exception.

Deceptive Depiction of a Candidate

The core target is synthetic media that falsely shows a real, identifiable candidate (or sometimes an election official) saying or doing something they did not, in a way a reasonable viewer could mistake for genuine. Statutes commonly use phrases like "materially deceptive" to separate convincing fakes from obvious edits.

Intent and Timing Requirements

Most laws require some level of intent, such as intent to deceive voters or to harm a candidate's chances, and most apply only within a pre-election window. These requirements narrow the laws and are part of how legislators try to keep them constitutional, since punishing all altered political imagery would sweep in protected speech.

Satire, Parody, and News Coverage Exemptions

Nearly every election deepfake law carves out satire, parody, and bona fide news reporting, and many exempt content that carries a clear "this is synthetic" disclaimer. These exemptions are not loopholes so much as constitutional necessities: parody and commentary are protected speech, a point that has driven several court challenges, discussed below.

Penalties and Enforcement

Penalties vary widely by state. Some statutes create criminal offenses (often misdemeanors), while many rely on civil remedies. A common and powerful tool is injunctive relief: a candidate can ask a court to order a deceptive deepfake taken down before it spreads further. Some laws also allow money damages.

Who can enforce these laws also varies. Depending on the state, an affected candidate may be able to sue directly, and state attorneys general or election officials may have enforcement authority. The FCC's robocall rule adds a federal enforcement layer when the deepfake is delivered by phone.

Enforcement is genuinely hard in practice. Deepfakes spread across state and national borders in minutes, anonymous creators are difficult to identify, and a court order to remove a clip often arrives after it has already gone viral. A fast takedown or correction can matter more than a later penalty, which is part of why platform reporting and detection are central to the response.

The First Amendment Problem

Election deepfake laws sit on contested constitutional ground. Political speech receives the strongest First Amendment protection, and courts have been skeptical of laws that restrict it, even to fight deception.

The clearest example is California. In October 2024, a federal judge preliminarily blocked most of California's AB 2839, a law that restricted "materially deceptive" election-related content, in the case Kohls v. Bonta. U.S. District Judge John A. Mendez wrote that the law acted "as a hammer instead of a scalpel" and likely violated the First Amendment by chilling humor and commentary, as reported by the nonpartisan Free Speech Center at MTSU. In August 2025, the same court issued a final ruling striking AB 2839 down as unconstitutional. Minnesota's election deepfake law has also faced litigation.

The takeaway is not that all these laws are doomed. Narrower disclosure requirements have generally fared better than broad bans, and courts are still drawing the line between policing fraud and suppressing speech. The practical point for readers: an "enacted" law is not always a currently enforceable one, so check a statute's litigation status before assuming it applies.

Real Cases: Election Deepfakes That Triggered the Law

A handful of documented incidents shaped this area of law. These are described factually, with official or established sources, and without any partisan framing.

The 2024 New Hampshire robocall is the landmark US case. An AI clone of President Biden's voice told New Hampshire voters not to vote in the primary; the FCC proposed a $6 million fine against the operative behind it and settled with the transmitting carrier for $1 million, per the FCC. The incident drove both the FCC's robocall ruling and a wave of state bills.

Internationally, a widely reported case in Slovakia involved fake audio of a political figure released days before a 2023 election, during a media blackout period when it was hard to rebut. It is a useful non-US contrast that shows why many US laws focus on the pre-election window. For more documented cases, see our deepfake examples page.

What to Do If You See a Political Deepfake

If you encounter a political clip that seems too inflammatory, too convenient, or just slightly off, here is a calm, nonpartisan way to respond.

1. Do not share it. Sharing, even to debunk it, spreads it. Pause before reposting.
2. Check the source. Look for the original poster and whether any credible outlet has reported the same thing. A real candidate statement is rarely a single anonymous clip.
3. Look for a fact-check. Reputable fact-checkers and newsrooms often address viral political media within hours.
4. Report it. Use the platform's reporting tools, and contact your state or local election officials if it concerns voting procedures or a candidate.
5. Verify it with a detector. Run the image, video, or audio through a detection tool to get an independent signal on whether it is likely synthetic.

FAQ

Are political deepfakes illegal?

There is no blanket federal ban. AI-voice robocalls are illegal under the FCC's ruling, and 30 states regulate election deepfakes, usually by requiring disclosure or restricting deceptive candidate content near an election.

Which states ban election deepfakes?

As of May 2026, 30 states have an election deepfake law, with Maryland the most recent, per Public Citizen. Approaches differ; see our deepfake laws by state table for your state.

Do AI political ads have to be labeled?

In many states, yes. Most state political deepfake laws use a disclosure model that requires a clear label on AI-generated or AI-altered campaign content. Federal rules are narrower and focus on robocalls.

Can a candidate sue over a deepfake?

Often yes. Many state statutes let an affected candidate seek a court order to remove a deceptive deepfake, and some allow damages. Existing defamation and false-light claims may also apply.

Is parody of a politician illegal?

Generally no. Parody and satire are protected speech, and nearly every election deepfake law explicitly exempts them. Courts have struck down laws seen as sweeping in this protected expression.

Conclusion: Verify Before You Vote or Share

Election deepfake laws are a fast-moving patchwork, not a settled rulebook. There is no blanket federal ban, the FCC has made AI-voice robocalls illegal, and 30 states regulate political deepfakes through bans, disclosure rules, or pre-election windows, while courts continue to test those laws against the First Amendment. For voters and journalists, the laws matter less in the moment than the habit behind them: slow down, check the source, and confirm before sharing.

If you saw a suspicious political clip and want an independent signal, you can check a video, image, or voice clip free, 50 detections a month, no card required. A detection report will not tell you whether a law was broken, but it can tell you whether the media is likely synthetic before you pass it on.

This article is general information, not legal advice. Last updated June 2026; statutes and cases cited were verified against official or established sources at that date. This page is on a standing 90-day re-review, monthly during major election windows.

Related reading