A video appears online showing a political candidate making a statement that was never made. A photograph of a teenager is altered to make her appear nude. Someone's face is placed into a sexually explicit video without permission. None of it actually happened, but the images can look convincing enough that the damage may begin long before anyone realizes they’re fake.
Artificial intelligence (AI) has made it easier to create realistic images, audio recordings, and videos of people saying or doing things they never said or did. Commonly known as "deepfakes," these digital creations have raised concerns ranging from election misinformation and fraud to harassment and nonconsensual intimate imagery.
The technology has advanced quickly enough that lawmakers are still deciding how existing laws should apply and where new protections are needed. Texas was one of the first states to specifically address deepfakes in elections, and it has since expanded its laws involving sexually explicit AI-generated content.
More recently, federal law has placed new responsibilities on online platforms when nonconsensual intimate images, including AI-generated ones, appear on their services.
For Texans, that means a deepfake is no longer simply a question of whether something online is real. Depending on how it is created and used, it can also have criminal and civil consequences.
What Is an AI Deepfake?
A deepfake is manipulated or artificially generated media designed to make it appear that a real person did or said something that didn’t actually occur.
Early deepfakes often involved videos in which one person's face was digitally placed onto another person's body. Generative AI has dramatically expanded what’s possible. Today's technology can create realistic photographs, imitate someone's voice, alter existing videos, or generate entirely new scenes.
The concern is not simply that fake media exists. Manipulated photographs have existed almost as long as photography itself. What has changed is the speed, accessibility, and realism with which convincing false material can now be created and distributed.
Research suggests that even identifying a video as a deepfake may not completely eliminate its influence. In a 2026 study, researchers conducted three experiments in which participants viewed deepfake videos depicting fictional people admitting to a crime or other misconduct. The researchers found that the videos continued to influence participants' judgments even after they were warned that what they were watching was fake.
That creates an unusual problem. A false image or recording may keep influencing someone's perception even after its authenticity is challenged.
Texas Was an Early State to Regulate Deepfakes in Elections
Political deepfakes are one of the most visible concerns surrounding artificial intelligence.
A realistic recording released immediately before an election could appear to show a candidate making an offensive statement, admitting wrongdoing, or taking a position the candidate never actually took. By the time the recording is disproven, millions of people may already have seen or shared it.
Texas addressed that possibility before generative AI became widely available. In 2019, Texas enacted Senate Bill 751, which amended the Texas Election Code to make it a criminal offense to create and distribute certain deceptive videos within 30 days of an election with the intent to injure a candidate or influence the result. The law defines a deepfake video as one created with the intent to deceive that appears to show a real person performing an action that didn’t occur.
Texas lawmakers considered broader proposals during the 2025 legislative session to address altered images, audio, and video in political advertising. Those measures didn’t become law.
The issue is hardly limited to Texas. States across the country have considered laws to govern deceptive AI-generated election material, often focusing on disclosure requirements or restrictions in the period immediately before an election.
That debate is likely to continue because political deepfakes pose a difficult legal balancing act. Lawmakers want to protect voters from deliberate deception while also accounting for political speech, satire, parody, and First Amendment protections.
Sexually Explicit Deepfakes Can Cause a Very Different Kind of Harm
Some of the most troubling uses of deepfake technology have little to do with politics.
AI tools can take an ordinary photograph of a real person and generate a sexually explicit image that never existed. Someone doesn’t need to have posed for an intimate photograph or recorded a private video to become the subject of one. This matters because the harm caused by an image doesn’t necessarily disappear simply because it's fake.
A fabricated sexual image can be sent to classmates or coworkers, posted on social media, used to harass a former partner, or employed as part of an extortion attempt. Once it begins circulating, copies may appear on other websites or accounts beyond the victim's control.
Texas originally made it a crime in 2023 to knowingly produce or electronically distribute certain sexually explicit deepfake videos without the depicted person's effective consent. Lawmakers significantly expanded those protections in 2025.
How Did Texas Change Its Deepfake Laws in 2025?
Senate Bill 441 broadened Texas law beyond deepfake videos to cover visual depictions created or altered through artificial intelligence, machine learning, software, or other technological means.
Under the revised law, knowingly producing or electronically distributing certain sexually explicit deepfake media without the depicted person's effective consent can be a Class A misdemeanor. The offense can rise to a third-degree felony in certain circumstances, including when the depicted person is younger than 18.
Texas law also addresses threats. Intentionally threatening to produce or distribute sexually explicit deepfake media to coerce, extort, harass, or intimidate someone can itself constitute an offense.
The 2025 law went further by creating or expanding civil remedies involving artificial intimate visual material. A person depicted in qualifying AI-generated intimate material may potentially pursue damages against someone who produces, solicits, discloses, or promotes the material without consent and with the intent to cause harm when the other requirements of the statute are satisfied.
Those provisions recognize something important about deepfake abuse. The consequences can extend beyond creating the image itself. Reputational damage, emotional distress, financial harm, and the continuing circulation of an image can remain long after the original post disappears.
What Responsibility Do Websites and Social Media Platforms Have?
The law surrounding deepfakes increasingly focuses not only on the person who creates an image but also on what happens after it’s posted online.
Texas law now requires websites and applications, including social media platforms, to provide an easily accessible system for requesting the removal of artificial intimate visual material.
Under the Texas statute, a website or application that discloses qualifying material may face liability if the depicted person requests its removal and the platform fails to remove it within 72 hours and make reasonable efforts to identify and remove known identical copies.
Federal law has added another layer of protection. The Take It Down Act was signed into law in 2025 and covers nonconsensual intimate images that are real as well as "digital forgeries," including material created or altered with artificial intelligence.
As of May 19, 2026, covered online platforms must provide a process for people to request removal of qualifying nonconsensual intimate images. After receiving a valid request, a covered platform generally must remove the material and known identical copies within 48 hours. The Federal Trade Commission (FTC) enforces those platform requirements and has begun warning companies that failure to comply can result in enforcement action.
Together, the Texas and federal laws reflect a significant shift in how policymakers approach online image abuse. The question is no longer limited to whether the creator of a deepfake can be punished. Lawmakers are also asking what responsibilities platforms have once they’re told harmful material is being distributed through their services.
What Can Someone Do If They Become the Target of a Deepfake?
The appropriate response depends heavily on what the deepfake depicts, who created it, how it was distributed, and whether it involves threats, extortion, sexual material, an election, or another form of misconduct.
One of the most important steps may be preserving evidence before content disappears. Screenshots, URLs, usernames, messages, dates, removal requests, and other information showing where and how the material was distributed can help document what occurred.
For people dealing with nonconsensual intimate imagery, the FTC's Take It Down service can help remove or stop the online sharing of intimate images, including AI-generated images, while allowing the images to remain on the person's device.
Texas law may provide additional criminal or civil remedies depending on the circumstances. The state's 2025 legislation also allows qualifying claimants in certain civil cases involving intimate visual material to use a confidential identity in court records and proceedings.
Deepfake cases can nevertheless be complicated. Identifying the person responsible may be difficult. A single image can cross multiple platforms and jurisdictions within hours. Different laws may apply to the person who created the material, to someone who distributed it, and to a platform that continued to host it after receiving notice.
Deepfakes Are Changing What It Means to Trust What We See Online
Artificial intelligence has made it possible to create useful images, videos, and other content in ways that would have seemed extraordinary only a few years ago. The same technology can also be used to convincingly place someone's face, voice, or identity into events that never occurred.
Texas lawmakers have already responded in areas where the potential harm is particularly significant, including elections and nonconsensual sexual imagery. Federal lawmakers have also begun requiring online platforms to respond more quickly when intimate images are distributed without consent.
Those laws will almost certainly keep evolving as technology changes.
For the public, the bigger change may be cultural. Seeing a photograph, hearing a recording, or watching a video can no longer always be treated as proof that an event actually happened. And when fabricated content is created to deceive, intimidate, exploit, or harm another person, the consequences can extend well beyond the screen.
Laws governing AI-generated content and online misconduct continue to evolve. If you believe you’ve suffered serious harm because of a deepfake or other digital content and have questions about your legal rights, contact The Cochran Firm Texas or call 1-800-843-3476 to discuss your situation with an attorney.
