This lesson explains compliance obligations for marketers. It is not legal advice. Talk to a lawyer before deploying AI-generated reviews, testimonials, or synthetic ad creative.
The rule that made fake reviews explicitly illegal
Fake reviews weren't new when generative AI arrived, but AI made them cheap to produce at scale. The FTC responded with a dedicated rule: on August 14, 2024, it announced a final rule banning fake and AI-generated consumer reviews, testimonials, and celebrity endorsements, effective October 21, 2024.
The rule prohibits reviews that misrepresent they're from someone who doesn't exist, someone who never used the product, or that misstate the reviewer's actual experience. It explicitly names AI-generated fake reviews as covered conduct, and it also bans buying reviews, suppressing negative ones through threats, and using fake social media engagement, follower counts, views, to inflate influence.
Penalties are real money: businesses face civil penalties currently set at $53,088 per violation, plus potential consumer redress. For a company generating thousands of fake reviews, that per-violation multiplier adds up fast.
Operation AI Comply and the Rytr case that got reversed
In September 2024, the FTC sued Rytr, an AI writing tool, as part of a sweep called "Operation AI Comply." The complaint alleged Rytr's review-generation feature let subscribers produce large volumes of detailed, specific-sounding reviews with no connection to any real product use, and put no cap on how many a subscriber could generate. In December 2024, the FTC issued a final order banning Rytr from selling any review or testimonial generation service for 20 years.
That order didn't last. In December 2025, the FTC reopened and set aside its own Rytr order, stating the original complaint didn't meet the legal bar and that the order "unduly burdens AI innovation" under the current administration's AI policy priorities.
This reversal doesn't undo the underlying fake reviews rule, that ban on AI-generated fake reviews and testimonials is still in force. It signals the FTC is now drawing a sharper line between banning deceptive output (fake reviews themselves) and penalizing the general-purpose AI tools that could theoretically be misused to create them.
The practical read for marketers: the target of enforcement is the deceptive review or testimonial you publish, not the AI tool you used to draft it. Using AI in your content pipeline isn't the violation; publishing AI output as if it were a real customer's unprompted experience is.
Deepfakes and impersonation get their own enforcement track
Separate from reviews, the FTC has been building out authority against AI-enabled impersonation. In February 2024, it proposed rule changes explicitly extending existing government- and business-impersonation protections to cover AI-generated impersonation of individuals, closing a gap where deepfaked voices, faces, or likenesses used in scams and marketing fell into a gray area.
For marketers, this matters most in two spots:
- Synthetic testimonials or endorser likenesses. Using an AI-generated voice or face styled after a real, identifiable person, celebrity or otherwise, without their consent risks both FTC action and separate right-of-publicity claims.
- AI voice-clone or deepfake ad creative. Ad campaigns using synthetic media that implies a real person said or did something they didn't cross from "creative" into "deceptive" the moment a reasonable viewer would believe it's authentic and unlabeled.
The safe pattern here isn't new, it's the same "reasonable consumer" deception test the FTC has always applied, just pointed at a new production method.
What to actually put in your AI content workflow
None of this requires abandoning AI tools in marketing, it requires treating AI output the same way you'd treat any claim you publish under your brand's name.
- Never publish AI-generated text as a customer quote or review unless it's disclosed as illustrative or synthetic, and ideally, just don't, real testimonials exist for a reason.
- Disclose synthetic media in ad creative when a reasonable viewer could otherwise believe a real person is speaking or endorsing your product.
- Keep a human review step before any AI-assisted review, testimonial, or endorsement content goes live, this is your main defense if the FTC asks how a fake review made it to your site.
- Watch the regulatory pace. 2025 alone saw a new rule take effect, a major sweep, and a reversal, this is one of the fastest-moving corners of ad law right now, not a settled one.
Treat AI as a drafting tool for your marketing copy, not a substitute for a real customer's voice, and you stay on the right side of a rule that's clearly still being tested in real time.