In this Law360 article, Laura-Kate Bernstein discusses compliance with the TAKE IT DOWN Act, a federal law that highlights growing concern about the rapid spread of nonconsensual intimate imagery, including AI-generated forgeries that can be produced and distributed on a large scale. As the FTC has expressed that violations of the act will be treated as violations of an FTC rule, the law leaves several questions unanswered.
“The statute applies only where imagery was published without consent, yet it offers little guidance regarding what evidence platforms should assess, how they should handle disputes over consent, or what level of technical analysis is required to determine whether an image is AI-generated or digitally manipulated.”
Please note that the original article is paywalled.
Read the original article here →
Unpacking The TAKE IT DOWN Act’s Compliance Ambiguities
The Tools to Address Known Exploitation by Immobilizing Technological Deepfakes on Websites and Networks, or TAKE IT DOWN, Act arrived with overwhelming bipartisan support, a compelling set of victim protection goals and unusually direct political momentum.1
Signed into law on May 19, 2025, the statute reflects mounting concern over the proliferation of nonconsensual intimate imagery, including artificial intelligence-generated “digital forgeries” that can be produced and distributed at unprecedented speed and scale.
One year later, the law’s platform obligations are now in effect, and the Federal Trade Commission has made it clear that enforcement will be a priority.
In May alone, the FTC published compliance guidance, issued a consumer alert and sent warning letters to major technology companies, including Amazon.com Inc., Apple Inc. and Meta Platforms Inc. — emphasizing that the agency expects covered platforms to comply with the law’s takedown requirements immediately.2[2]
The FTC’s May guidance states that violations of the TAKE IT DOWN Act will be treated as violations of an FTC rule, potentially exposing companies to civil penalties exceeding $53,000 per violation. FTC Chairman Andrew Ferguson has likewise emphasized that the agency stands “ready to monitor compliance, investigate violations, and enforce the TAKE IT DOWN Act.”
The speed and seriousness of the rollout are notable. But the most significant feature of the statute may ultimately be its breadth. Although much of the public conversation around the law has focused on major social media companies, the statutory definition of “covered platform” sweeps far more broadly than many businesses may appreciate.
At the same time, the law leaves unresolved several critical operational questions, including what level of investigation platforms must undertake before removing allegedly unlawful content. Those ambiguities matter because the statute’s incentives strongly favor rapid removal, even where the underlying claim may be disputed or difficult to verify.
The Statutory Framework
The TAKE IT DOWN Act contains two principal components. First, it creates criminal penalties for knowingly publishing or threatening to publish nonconsensual intimate visual depictions, including AI-generated intimate imagery that is “indistinguishable from an authentic visual depiction.” Second, and more consequential for most technology companies, Section 3 of the law establishes a notice-and-takedown regime enforced by the FTC.
Under that regime, covered platforms must provide a clear and conspicuous process allowing individuals or their authorized representatives to request removal of nonconsensual intimate imagery appearing on the platform. Once a valid request is submitted, the platform must remove the reported content — and make reasonable efforts to remove known identical copies — within 48 hours.
On paper, the framework may resemble other familiar notice-and-takedown systems. In practice, however, the statute departs from prior intermediary liability frameworks in important ways. The law imposes an extraordinarily short turnaround time. It also applies to a category of content that is often highly contextual, emotionally charged and difficult to verify.
And unlike copyright claims under the Digital Millennium Copyright Act, the TAKE IT DOWN Act does not establish a detailed counternotice process for users whose content is removed.
What Qualifies as a Covered Platform
Perhaps most importantly, the law defines “covered platform” expansively. The statute applies to any website, online service, online application or mobile application that “primarily provides a forum for user generated content,” including messages, videos, images, games and audio files.
The FTC’s May guidance reinforces that breadth, stating that the law may apply to “social media, messaging, image or video sharing, and gaming platforms,” among others.
That definition potentially reaches far beyond traditional social networking companies. Messaging services, community forums, gaming platforms with user chat functions, collaboration tools, creator economy products, cloud-sharing services, dating applications and niche consumer communities may all fall within the statutory language if they meaningfully facilitate user-generated content.
The first wave of FTC warning letters likewise spanned a wide range of services, including communications products, community forums, website-hosting providers and image-sharing services, suggesting the agency does not view the law as limited to a narrow category of consumer social media companies.
Implementation Challenges and Compliance Ambiguities
For many companies, the more difficult questions involve implementation. Covered platforms must provide a clear process for individuals to request removal of nonconsensual intimate imagery, and, upon receiving a valid request, remove the content and make reasonable efforts to remove known identical copies within 48 hours.
But the statute does not clearly explain what makes a request valid. Is a request valid simply because it includes the required information? Or are platforms expected to independently review the flagged material and determine whether it actually qualifies as nonconsensual intimate imagery? That distinction matters.
If procedural completeness alone triggers removal obligations, platforms may feel compelled to remove content even where a report is facially implausible or plainly incorrect. But if platforms are expected to substantively evaluate flagged material, the compliance burden becomes far more complicated.
The statute applies only where imagery was published without consent, yet it offers little guidance regarding what evidence platforms should assess, how they should handle disputes over consent, or what level of technical analysis is required to determine whether an image is AI-generated or digitally manipulated. The law likewise provides little direction for handling requests involving potentially newsworthy, political, satirical or otherwise protected expression.
Those challenges are magnified by the statute’s 48-hour deadline. Large platforms may have sophisticated trust and safety teams and automated moderation tooling, but many potentially covered services do not. Smaller companies may lack personnel capable of assessing consent disputes, authenticating images, identifying AI manipulation or evaluating contextual defenses on such an accelerated timeline.
Safe Harbor and Incentives for Over-Removal
The law’s safe harbor further complicates the analysis. It protects platforms that in good faith remove or disable access to material claimed to constitute nonconsensual intimate imagery where the facts or circumstances make the unlawful nature of the content apparent. That language appears to contemplate at least some degree of substantive review by platforms themselves, rather than merely checking whether a request is procedurally complete.
At the same time, the safe harbor creates strong incentives to err on the side of removal. Violations of the law may expose companies to substantial FTC penalties, while the risks associated with erroneous takedowns are comparatively limited. Platforms therefore may have strong incentives to remove content first and investigate later, particularly where AI-generated or digitally altered imagery is involved.
The FTC’s guidance also emphasizes that platforms must make “reasonable efforts” to identify and remove known identical copies of reported content. For larger services, that may involve automated detection and matching tools already used in existing trust-and-safety workflows. Smaller companies, however, may struggle to deploy systems capable of reliably identifying duplicate or slightly modified content at scale.
Looking Ahead
None of this diminishes the very real harms caused by nonconsensual intimate imagery or the understandable policy goals underlying the statute. But fast-moving takedown systems inevitably create tension between victim protection and speech interests. The TAKE IT DOWN Act places platforms directly in the middle of that tension while providing limited guidance regarding how to resolve difficult edge cases.
For now, the FTC’s message is clear: Platforms should move quickly, build removal systems immediately and prioritize compliance. But until courts or regulators provide additional clarity, many companies will be left navigating a statutory framework that is simultaneously expansive, urgent and uncertain.
Reprinted with permission from Law360. © 2026 Portfolio Media, Inc. Further duplication without permission is prohibited. All rights reserved.
- https://www.congress.gov/119/plaws/publ12/PLAW-119publ12.pdf. ↩︎
- https://www.ftc.gov/business-guidance/resources/complying-take-it-down-act, https://consumer.ftc.gov/consumer-alerts/2026/05/what-will-ftcs-enforcement-take-it-down-act-mean-you, https://www.ftc.gov/news-events/news/press-releases/2026/05/ftc-chairman-ferguson-advises-companies-comply-take-it-down-act. ↩︎