How Do Ai Deepfake Laws Differ Across U.S. States?

Deepfake rules start at the state level. States hold broad power over harassment, privacy, and defamation, so most early deepfake statutes came from state legislatures rather than Congress. That is why coverage looks like a patchwork: one state may jail a creator while another offers only a civil claim for the same act.

Two design choices drive most of the difference. The first is whether a statute lets victims sue directly, not just refer the matter for prosecution. The second is whether it targets a narrow harm, such as sexual images or election fakes, or reaches deception more broadly. Free-speech limits also slow broader laws, since courts protect satire, parody, and political commentary.

New York gives victims more than one path. The state's 2023 deepfake law added digitally created images to its crime of unlawful dissemination of an intimate image and created a private right of action, so a victim can press charges, sue, or do both. The table below shows how that approach differs from other leading states.

Federal law now sits on top of these state rules. The TAKE IT DOWN Act of 2025 makes nonconsensual intimate deepfakes a federal offense and requires covered platforms to remove reported content, and known copies, within 48 hours of a valid request. The FTC began enforcing that removal duty in 2026.

The federal statute sets a national floor, but it does not replace state law. It centers on intimate images, so it leaves election fakes, financial scams, and other non-sexual deepfakes to state statutes and general civil claims. In practice, a New York victim often uses the federal takedown right to pull content down fast, then relies on New York law to sue or press charges.

New York's intimate-image law covers sexual content, so it does not reach every fake. A video that shows you confessing to a crime or endorsing a product causes real harm, yet it falls outside that statute. Victims of non-sexual fakes turn to laws that apply no matter the technology.

A fake that spreads a false and damaging claim can support a defamation lawsuit, and unauthorized commercial use of your face or voice can trigger right-of-publicity claims under New York Civil Rights Law §§ 50 and 51. Fraud is the other main route. Cloned audio that authorizes a transfer, or a synthetic identity used to open accounts, can bring identity theft charges under state law and federal wire-fraud charges once the scheme crosses state lines.

Because protection depends on the type of harm, start by identifying what the deepfake shows and where it spread. From there, you can act on more than one front at the same time:

• Ask the platform to remove the content; under the TAKE IT DOWN Act, covered sites must act within 48 hours of a valid request.
• Report the incident to the NYPD, or to the FBI's Internet Crime Complaint Center when fraud or interstate conduct is involved.
• Ask a New York attorney whether a state statute, a defamation claim, or a cyberstalking claim best fits your facts.

Can you go to jail for making a deepfake in New York?

Yes. Sharing a nonconsensual intimate deepfake can be charged as unlawful dissemination or publication of an intimate image, a class A misdemeanor under New York Penal Law § 245.15. Non-sexual deepfakes used to defraud or harass someone can bring other criminal charges instead.

Do New York deepfake laws apply if the person who made it lives in another state?

They can. New York courts may take jurisdiction when the harm reaches a New York resident, and federal law applies across state lines. Enforcement gets harder when the creator is anonymous or overseas, which is why many victims pursue platform removal and civil claims at the same time.

Does New York's deepfake law cover AI voice cloning?

Not directly. The state's intimate-image law targets visual depictions, so a cloned voice usually falls outside it. A voice deepfake used to steal money or impersonate someone is generally handled under fraud, identity-theft, or harassment laws instead.