Vermont Deepfake Laws: AI Images, Voice Cloning & Penalties (2026)

Vermont has two active deepfake laws: a sexual-image statute covering AI-generated content since June 2024 and an election-media disclosure law signed in March 2026. No state law explicitly criminalizes wholly AI-generated child sexual abuse material (CSAM) without a real minor, but federal law fills that gap. Vermont has not yet passed a voice-cloning or digital-likeness statute.

Is It Illegal to Make a Deepfake of Someone in Vermont?

It depends on the type of deepfake. Vermont law targets two categories directly: sexually explicit or intimate imagery of adults, and AI-generated content in campaign advertising. A deepfake that falls into neither bucket is not criminalized under current state law, though it may still create civil liability under other theories (defamation, false light) or trigger federal statutes.

The three buckets for any state analysis are: (1) sexual and intimate deepfakes targeting adults and the separate question of AI-generated imagery involving minors, (2) election and political deepfakes, and (3) voice cloning and digital likeness. Vermont has addressed the first two but not the third.

For a broader look at how Vermont regulates artificial intelligence across sectors, see Vermont AI Laws, which covers algorithmic accountability, automated decision-making, and AI in employment, distinct from the deepfake-specific statutes addressed here.

Sexual and Intimate Deepfakes

Vermont criminalizes nonconsensual disclosure of intimate visual images under 13 V.S.A. s. 2606, first enacted in 2015 and amended by Act 161 effective June 6, 2024. The 2024 amendment explicitly extended the definition of "visual image" to include images "created or altered by digitization," which means AI-generated deepfakes are squarely covered regardless of whether the underlying photograph is real.

To be criminal, the disclosure must be knowing, the image must show an identifiable person who is nude or engaged in sexual conduct, the person depicted must not have consented, the discloser must act with intent to harm, harass, intimidate, threaten, or coerce the person depicted, and the disclosure must be one that would cause a reasonable person to suffer harm. The statute also makes clear that consenting to the recording of an image does not, by itself, mean consenting to its disclosure.

Penalties scale with motive. A base violation carries up to 2 years imprisonment and a fine of up to $2,000. If the image is distributed with intent to obtain financial profit, the maximum rises to 5 years imprisonment and a $10,000 fine.

Victims also have a civil remedy: s. 2606 authorizes a private lawsuit for equitable relief, including temporary restraining orders, preliminary injunctions, and permanent injunctions requiring the defendant to stop distributing the image.

AI-Generated CSAM: A Meaningful Gap

Vermont has not explicitly updated its child sexual exploitation statutes to cover wholly AI-generated imagery of minors. Vermont's existing CSAM laws (13 V.S.A. s. 2821 et seq.) are written around the sexual performance or exploitation of real children. A realistic AI-generated image depicting a child in a sexually explicit situation, with no real minor involved in its creation, does not clearly fall within those state provisions.

This is a meaningful legal gap shared by roughly five states nationwide. It does not mean such material is legal in Vermont. Federal law under 18 U.S.C. 2256(8)(B) (PROTECT Act, 2003) expressly covers "digital image, computer image, or computer-generated image that is, or is indistinguishable from, that of a minor engaging in sexually explicit conduct." Anyone who creates, possesses, or distributes such material in Vermont faces federal prosecution even in the absence of a matching state charge.

The federal standard is objective: if an ordinary person viewing the image would conclude it depicts a real minor in sexually explicit conduct, the image is covered. Pure cartoons and hand-drawn illustrations are excluded, but photorealistic AI-generated output is not.

Election and Political Deepfakes

Vermont enacted Act 75 (S.23) on March 5, 2026, becoming one of roughly two dozen states with a law specifically targeting AI-generated content in campaign advertising. The law addresses the risk that synthetic media can deceive voters by making candidates appear to say or do things they never did.

Act 75 requires anyone who publishes deceptive and fraudulent synthetic media to include a disclosure stating: "This media has been manipulated or generated by digital technology and depicts speech or conduct that did not occur." The duty applies when the material is published within 90 days of an election in Vermont and the person knows the media is deceptive and fraudulent, defined as a realistic AI depiction that injures a candidate's reputation or gives voters materially false information to influence an election.

Penalties scale with intent and history. A knowing violation carries a fine of up to $1,000. The cap rises to $5,000 if the violation involved intent to cause violence or bodily harm, to $10,000 for a repeat violation within five years, and to $15,000 when both factors are present. A misrepresented candidate may also seek injunctive relief under 17 V.S.A. 2033(b), and the Attorney General or a State's Attorney enforces the law. The law took effect on passage, meaning it applied to the 2026 election cycle.

A First Amendment caveat applies here as it does in other states: on August 29, 2025, a federal court held that California's parallel AI election-deepfake law (AB 2839) failed strict scrutiny in its entirety, finding no portion of it severable, and permanently enjoined the California Attorney General and Secretary of State from enforcing it against the named plaintiffs in that case (Kohls v. Bonta, 797 F. Supp. 3d 1177 (E.D. Cal. 2025)). That injunction runs to those plaintiffs rather than wiping the statute off the books statewide, but the court's First Amendment reasoning reaches any similarly drafted law. Vermont's Act 75 targets disclosure rather than outright prohibition, which is generally viewed as constitutionally safer, but the legal landscape for election-deepfake statutes remains unsettled nationwide.

AI Voice Cloning and Digital Likeness

Vermont has no statute equivalent to Tennessee's ELVIS Act (Tenn. Code Ann. 47-25-1101 et seq., eff. July 1, 2024), which was the first state law to extend right-of-publicity protections to AI voice simulations. The ELVIS Act is the national reference point for this category of legislation.

Under current Vermont law, an AI voice clone used in a commercial product, advertisement, or political message without the subject's consent is not expressly criminalized. A victim may attempt civil claims under common-law right of publicity, misappropriation of likeness, or false endorsement under the Lanham Act, but Vermont has not codified voice-clone protections.

The proposed federal NO FAKES Act (S.4591, 119th Congress) would create a national right of publicity covering AI voice and likeness replicas. As of June 2026 it has not passed either chamber and is proposed only; it is not law.

If you are a performer, broadcaster, or public figure concerned about AI voice cloning in Vermont, the most relevant current protections are the FTC's impersonation rule (16 CFR Part 461) covering business and government impersonation, and the FCC's ruling (FCC 24-17) making AI-generated voices in robocalls illegal under the TCPA.

Federal Law That Applies in Vermont

Several federal laws protect Vermont residents regardless of state coverage.

The TAKE IT DOWN Act (Public Law 119-12, signed May 19, 2025) is the first federal law criminalizing nonconsensual intimate visual depictions, expressly including AI deepfakes described in the statute as "digital forgeries." It covers both adults and minors. The penalty is up to 2 years imprisonment, rising to 3 years when minors are involved. Platforms must remove flagged intimate images within 48 hours of receiving notice from a victim; that compliance deadline took effect May 19, 2026, and is enforced by the FTC.

Because Vermont's s. 2606 already covers adults, the TAKE IT DOWN Act adds the most value in the minor-deepfake context and as a platform-removal mechanism. A victim who cannot get a state criminal prosecution moving can use the federal 48-hour takedown process to get content removed from major platforms quickly.

For AI-generated CSAM, 18 U.S.C. 2256(8)(B) applies as described above under the minor-image gap section.

The proposed DEFIANCE Act (S.1837, 119th Congress) would create a federal civil cause of action for victims of sexual deepfakes with liquidated damages up to $150,000 ($250,000 if the conduct involved actual or attempted sexual assault, stalking, or harassment). The 118th-Congress version passed the Senate in July 2024 but died in the House. The 119th-Congress version passed the Senate by unanimous consent on January 13, 2026, and is pending in the House as of June 2026. It is not law. For background, see our coverage of the DEFIANCE Act and victims' right to sue.

The FTC's impersonation rule (16 CFR Part 461) prohibits using AI to impersonate government entities or businesses. The individual-impersonation extension remains an unfinalized proposed rulemaking. The FCC's ruling (FCC 24-17) makes AI-generated voices in robocalls illegal under the TCPA, which was triggered by the fake-Biden AI robocall in New Hampshire's 2024 primary.

What Victims Can Do

If you are a victim of a sexual deepfake in Vermont, several pathways are available.

File a criminal complaint with local police or the Vermont Attorney General's office citing 13 V.S.A. s. 2606. The statute covers both the creator and anyone who redistributes the image knowingly. Bring documentation of where the image appeared and any evidence of who created or shared it.

File a civil lawsuit under s. 2606 for an injunction. An emergency temporary restraining order can require a defendant to stop distributing the image while the case proceeds. If the deepfake was created for financial profit, ask your attorney about whether the enhanced penalty provision strengthens your civil position.

Use the TAKE IT DOWN Act's 48-hour platform-removal rule. Under Public Law 119-12, major platforms must remove flagged nonconsensual intimate images within 48 hours of a victim's notice. This is the fastest route to getting content off public platforms without waiting for criminal prosecution.

For election deepfakes, report missing synthetic-media disclosures to the Vermont Attorney General or your county State's Attorney, who enforce Act 75. A candidate targeted by deceptive synthetic media may also seek an injunction directly under 17 V.S.A. 2033(b).

For AI voice cloning used in robocalls, a complaint can be filed with the FCC.

Vermont residents dealing with deepfake content should also explore whether Vermont data privacy laws provide any additional basis to demand deletion or restrict further processing of their image or biometric data.

For context on how Vermont handles related recording and surveillance issues, see Vermont recording laws.

Penalties at a Glance

Disclaimer: This page provides general legal information about Vermont deepfake and AI-generated image laws, not legal advice. AI-related laws are changing rapidly; statutes described here may have been amended after publication. If you have been harmed by a deepfake or face a legal claim, consult a licensed Vermont attorney.

Sources

This page cites Vermont primary statutes and federal law; citations appear below.

Deepfake & AI Voice Cloning Laws by State

More Vermont Laws

• Vermont AI Meeting Recording Laws
• Vermont Data Privacy Laws
• Vermont Recording Laws
• Vermont Sexting Laws