AI makes it trivially easy to write a superlative, a statistic or a glowing testimonial. Advertising law hasn't changed: claims need proof, and endorsements need to be real.
Claims need proof before you publish
In the US, the FTC's long-standing advertising substantiation policy says advertisers must have a reasonable basis for objective claims before they are published. "Best", "#1", "fastest", "clinically proven", "9 out of 10" — if an AI tool writes them, someone still needs the evidence on file. Other markets have similar rules.
Fake reviews are explicitly banned
The FTC's final rule on reviews and testimonials, announced August 14, 2024, bans reviews and testimonials that misrepresent that they are by someone who doesn't exist — the FTC names AI-generated fake reviews as an example — or by someone who didn't actually use the product. It also lets the FTC seek civil penalties. Buying such reviews when you knew or should have known they were fake is covered too.
Where AI-made content commonly breaks the rules
- Invented statistics in drafts ("studies show 80%...").
- Made-up customer quotes used as "examples" that end up live.
- AI personas presented as real customers or experts.
- Comparative claims about competitors that no one checked.
Simple controls
- An instruction in every AI tool: never invent facts, figures, quotes, names, reviews or testimonials; mark [FACT NEEDED].
- An automated check that flags superlatives and figures without a source.
- A claims register: each claim, its evidence and its expiry date.
- Legal sign-off for high-risk claims (health, price, performance, comparisons).