How state lawmakers are acting to stop the harm of AI-generated deepfakes
Many people first became aware of AI-generated deepfakes in Oct. 2023, when actor Tom Hanks issued a warning about a scam using his deepfaked image and voice to grift unsuspecting consumers.
The proliferation of deepfakes online has driven many states to pass legislation banning or limiting the artificial intelligence-driven technology, particularly if it’s being used to maliciously represent someone against their will.
Deepfakes are highly realistic yet false images, audio recordings or videos created using the voice or image of someone – typically without their consent and often doing or saying things they would never do or say.
Most states passed laws pre-internet that made it a crime to impersonate someone, particularly law enforcement, lawyers, doctors, or government officials. It’s possible those laws could apply to deepfakes. But, since 2019, many states have passed anti-deepfake legislation to strengthen existing protections, according to the National Conference of State Legislatures.
The pace of legislation increased significantly last year, as more and more examples of harmful deepfake deception made news. Through the end of 2024, at least 50 new laws mentioning or directly dealing with AI-generated deepfake imagery had been enacted in states across the country.
38 deepfake bills filed in 18 states this year
So far in 2025, legislators in 18 states have introduced 38 bills that deal with AI-generated deepfake imagery.
The Transparency Coalition is alarmed and horrified by the proliferation of these harmful images, especially as they impact the lives of our kids. Over the past year we’ve been inspired by the action we’ve seen taken by local state lawmakers to halt the harm. Our organization’s focus is on other aspects of AI — in part because we see so much attention and action already given to deepfakes —but we fully support legislation that appropriately protects individuals, and especially kids, from this new and too-easily abused technology.
To keep our readers up to date on the issue, TCAI has created a bill tracker specifically tuned to deepfake legislation introduced this year. It will remain live and updated daily through 2025.
TCAI 2025 Bill tracker: deepfake bills
new laws on abusive deepfakes in 17 states
At least 17 states have already passed legislation that specifically bans the use of artificial intelligence to intimidate, bully, threaten or harass a person through social media sites, email or other electronic or online communications.
The following states have barred such abusive deepfakes: California, Connecticut, Florida, Hawaii, Illinois, Louisiana, Massachusetts, Mississippi, New Jersey, New York, North Carolina, Oklahoma, Rhode Island, Texas, Utah, Washington and Wyoming.
Not all of the existing laws focus on AI-created deepfakes. Many apply more broadly to fraudulent, manipulated audio, computer-generated video or images that depict people falsely without their consent.
Many of the laws focus on sexually explicit or pornographic video or images, with some states - including Georgia, Hawaii, New York and Virginia - enhancing existing revenge porn laws to include deepfakes. Other states, such as Florida, Minnesota, and Texas, drafted new statutes and amendments to cover sexually explicit deepfakes.
New laws on the use of deepfakes in political ads
Other states, including Minnesota and Texas, have focused on prohibiting distribution of deepfakes that could harm a political candidate’s reputation or mislead a voter in making a decision to vote for or against a candidate.
However, some, including California, Michigan, and Washington state, have exemptions or provide an affirmative defense for using deepfakes if a disclaimer is displayed on the political ad. These laws vary from state to state, in terms of whether a specific malicious intent is required in order to qualify as a violation of the law and the resulting liability. Some states make it a crime, while others, including California, Florida, Illinois, and Minnesota, only attach provisions for civil or injunctive relief.
Some states updating existing ‘right of publicity’ laws
One final way that states are limiting the use of deepfakes is via “right of publicity” laws, or laws that protect the intellectual property of a person’s name, image, and likeness - including their voice, signature or photographs – from being used by others for commercial benefit.
For example, New York updated its right of publicity statute in 2021 to extend those rights to celebrities and performers even after they die.
Tennessee recently passed the Ensuring Likeness Voice and Image Security Act (ELVIS Act), which expands the state’s right of publicity law to protect against the use of a person’s voice, or a simulation of that voice. The ELVIS Act received significant support from the music industry.
deepfake laws in the age of ai
At least 50 deepfake-related bills have been enacted through the end of 2024. Some highlights:
Alabama: It’s now a crime to knowingly create, record, or alter a private image when the depicted individual has not consented to the creation, recording, or alteration and the depicted individual had a reasonable expectation of privacy.
California: The state now allows people to report digital identity theft to any social media platform where a deepfake appears, and requires the social media platforms to permanently block a reported instance of digital identity theft from being publicly available on that platform.
Florida: Now requires certain political advertisements, electioneering communications, or other advertisements that use deepfakes to include a disclaimer. It establishes criminal and civil penalties for violating the law.
Iowa: Barred the sexual exploitation of a minor via the creation, adaptation, or modification of an image or video falsely showing an identifiable minor engaged in a prohibited sexual act or the simulation of a prohibited sexual act.
Louisiana: Made it a crime to disseminate or sell AI-created images of another person.
South Dakota: Expanded the definition of child pornography to include AI-generated child pornography.
Tennessee: The use of a person’s name, image, voice or likeness, in any medium in any matter, is now a protected personal property right.
Utah: Updated the definition of counterfeit intimate image to include AI- or computer-generated images.
learn more:
