An AI political ad is not automatically a deepfake. “Political ad” describes a communication’s purpose, sponsor, and distribution; “deepfake” describes realistic media that falsely depicts a person or event. The categories can overlap: a political ad may use a deepfake, but it may also use AI for routine edits or imagined visuals that do not create a realistic false depiction. Whether a disclosure is required depends on the law, platform, medium, and circumstances.
What makes something a political ad, a deepfake, or manipulated media?
These terms describe different aspects of a piece of content. A communication may fit one category, more than one, or neither under a particular law or platform policy.
As an Amazon Associate I earn from qualifying purchases.
- Political ad: A communication treated as political under the applicable law or platform rules. The Federal Election Commission’s definition of public communications covers general public political advertisements in formats such as broadcast, cable, satellite, print, outdoor, certain mass mailings and phone banks, as well as paid placements on another party’s website, digital device, application, or ad platform. That regulatory scope is not necessarily the same as everyday usage.
- Deepfake: In common explanatory usage, realistic media created or altered to falsely depict a person’s face, voice, speech, actions, or an event. There is no single nationwide legal definition established by the sources discussed here; the precise definition depends on the law being applied.
- Other manipulated media: Edited or generated image, audio, video, or context. Some changes are routine and do not materially affect the message; others create a misleading impression. The fact that AI was involved, by itself, does not establish that the result is deceptive or legally a deepfake.
Arizona’s statute, for example, defines synthetic media as an image, audio recording, or video recording of a person’s appearance, speech, or conduct created or intentionally manipulated with digital technology to make a realistic but false depiction. That is a definition for that statute, not a universal one.
Recommended Free Tools
How to tell the categories apart
Start by separating the communication’s political context from what its media depicts. Then consider the alteration, its effect on the claim, and the rules that apply to its distribution. No single test governs every U.S. law and platform policy.
#1 Best Overall
- Identify the communication and sponsor. Is it a political communication under the relevant rules, and who paid for or authorized it?
- Describe the alteration precisely. Was a real person made to appear to say or do something they did not, or was a realistic event depicted that never happened? Or was the change a routine edit such as cropping or color correction?
- Ask whether the change affects the message. An alteration that changes the substantive claim or creates a realistic false impression is different from one inconsequential to the ad’s claims.
- Check the applicable conditions. Some rules turn on intent, knowledge, election timing, medium, or the identity of the depicted person. Also check exceptions and the law or platform policy for the place and channel where the ad appears.
These questions help organize the issue, but they are not a universal legal checklist. The specific tests below differ materially.
What U.S. rules and platform policies say
The examples below illustrate different approaches, not a complete survey of U.S. federal, state, or local requirements. They reflect materials accessed or issued by October 7, 2026.
Rank #2
| Authority | What it addresses | Disclosure or rule |
|---|---|---|
| Federal Election Commission (FEC) | Existing federal fraudulent-misrepresentation provisions and disclaimers for covered communications | No blanket federal requirement to label every political ad that uses AI. The FEC says existing provisions are technology-neutral and will be applied case by case. |
| Arizona | A defined kind of deceptive and fraudulent deepfake involving a candidate on the ballot, within 90 days before an election | A clear and conspicuous AI-generated-content disclosure is required when the statute’s conditions are met; the statute includes exceptions. |
| Florida | Covered political communications using generative AI to depict a real person doing something that did not occur, subject to specified intent conditions | A prominent disclaimer is required under the 2025 statute, with presentation rules varying by medium. |
| Google election-ad policy | Synthetic or digitally altered images, video, or audio that inauthentically depict real or realistic-looking people or events | Advertisers must select an “Altered or synthetic content” checkbox; Google generates an in-ad disclosure for some formats, while other formats require the advertiser to provide one. |
Federal Election Commission: existing rules, not a universal AI label
On September 19, 2024, the FEC voted not to open a new rulemaking on AI use in campaign ads. Its September 27, 2024 summary explains that existing fraudulent-misrepresentation provisions are technology-neutral and apply regardless of the technology used, including AI-assisted media. The Commission said it would continue to assess cases individually. That position is not a general requirement to disclose every use of AI.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteThe FEC separately describes payer and authorization disclaimers for covered communications. For television communications subject to the “stand by your ad” provision, its guidance says a written disclaimer must appear at the end for at least four seconds, use reasonable color contrast, and occupy at least four percent of the vertical picture height. Those specifications concern sponsor disclaimers, not a general AI-content label.
The Commission reported receiving more than 2,000 comments after seeking public comment on a 2023 petition for rulemaking on deceptive AI-produced campaign content. That figure describes comments received, not a measure of how common AI ads or deepfakes are.
Arizona: a narrow statutory deepfake rule
Arizona Revised Statutes § 16-1024 addresses a creator who, within 90 days before an election, knowingly creates and distributes a deceptive and fraudulent deepfake of a candidate on the ballot. The law’s definition involves a realistic but false depiction, knowledge that it is false, intent to injure the candidate’s reputation, and intent to mislead a reasonable person about what someone said or did. When the statutory conditions apply, the law requires a clear and conspicuous disclosure that the media includes AI-generated content. The statute exempts satire or parody and interactive computer services.
Florida: a prominent disclaimer for specified generative-AI content
Florida Statutes § 106.145 (2025) applies to a political ad or other covered political communication containing generative-AI content that appears to depict a real person doing something that did not occur, when created with intent to injure a candidate or deceive about a ballot issue. The required wording is: “Created in whole or in part with the use of generative artificial intelligence (AI).” The statute sets different presentation rules for print, video, online text or graphics, audio, and graphic communications.
Google: policy scope depends on the ad format
Google’s election-ad policy includes synthetic or digitally altered images, video, or audio that inauthentically depict real or realistic-looking people or events. Its examples of content in scope include making a person appear to say or do something they did not, or generating a realistic event that did not happen. It excludes changes inconsequential to the ad’s claims, with examples such as resizing, cropping, color or brightness correction, defect correction, and some background edits. The advertiser’s checkbox and disclosure duties vary by format; Google says its own disclosure does not replace disclosures required by law.
Best Value
- Used Book in Good Condition
Google’s requirements are a platform policy, not a general legal standard or a description of every platform’s rules.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Why an AI disclosure and a sponsor disclaimer are different
A sponsor disclaimer answers who paid for or authorized this communication? An AI or synthetic-content disclosure answers was the media generated or altered in a way covered by this rule? Those are separate questions, so one label does not automatically satisfy the other. Which disclosures are needed depends on the communication and the applicable rules.
What manipulated political media does—and does not—show about election security
An October 18, 2024 FBI and CISA advisory warned that generative-AI tools lowered the barrier for foreign malicious actors to run more sophisticated influence campaigns. It described tactics that could work together, including synthetic-media messages, deepfake video and audio, inauthentic articles, spoofed websites, paid influence, messaging channels, and fake online personas.
The same advisory said the agencies had no information that malicious cyber activity had compromised election infrastructure during the 2024 U.S. election cycle, prevented eligible voters from voting, affected ballots, or disrupted timely counting or reporting. The advisory concerns that election cycle and distinguishes influence operations from compromise of voting systems; evidence of one should not be presented as proof of the other.
What to check before drawing a conclusion about a specific ad
- Jurisdiction: Which federal, state, or local rule applies? Arizona and Florida are examples, not a nationwide standard.
- Timing and intent: Does the law specify an election window, knowledge, or a particular intent?
- Content and effect: Is a real or realistic person or event falsely depicted, and does the change alter the ad’s substantive message?
- Distribution: What medium and platform carried it, and what format-specific rules apply?
- Exceptions and current status: Does an exception apply, and have effective dates, amendments, or court decisions changed the rule?
For an actual campaign or compliance decision, consult the current law and applicable platform policy for the relevant jurisdiction and format; these examples are not individualized legal advice.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
