The Disclosure Compliance Audit
Objective: Given three draft social posts from a proposed virtual-persona awareness campaign, identify every FTC Section 5 and state-law disclosure defect, and distinguish real violations from cosmetic non-issues.
You're reviewing draft posts for Concord Biotech's patient-awareness division, the Ahmedabad-headquartered fermentation-based API manufacturer that IPO'd in 2023. Its agency has drafted a virtual persona, 'Dr. Aria,' to post general health-awareness content ahead of legal sign-off.
Audit three draft posts against the FTC's per-post disclosure requirement for synthetic personas and New York's AI Transparency in Advertising Act, flag every defect, and don't flag stylistic choices that aren't actually violations.
Before you start
What you'll need
Free path (everything below is enough to finish)
Free and gives legal a sortable record before sign-off
Free tier supports a shared, linkable checklist the agency and legal can both edit
The process
Specimens to review
Legal asks you to flag any disclosure problems with this draft before it's approved. What's wrong?
Draft post 1, Dr. Aria's account: 'Did you know most people don't realize how their medication is actually made? I dove into the science behind API manufacturing this week and it's honestly fascinating. Follow for more!' No disclosure of any kind, no mention that Dr. Aria is an AI-generated persona, no sponsorship label.
Specimen: synthetic, realistic
This draft has a sponsorship tag already. Is that enough?
Draft post 2, Dr. Aria's account: 'API manufacturing quality standards explained in 60 seconds. #ad' A single small hashtag reading '#ad' appears at the end of the caption; nothing else in the post or the persona's bio mentions AI, synthetic, or virtual.
Specimen: synthetic, realistic
The bio already discloses the AI persona and sponsorship. Does that cover every post?
Draft post 3, Dr. Aria's account: Bio: 'AI-generated educational persona, sponsored by Concord Biotech.' Caption of the post itself: 'Three things nobody tells you about API purity testing.' No disclosure text appears in the post caption or the video itself.
Specimen: synthetic, realistic
Final deliverable
A disclosure compliance audit memo flagging all three draft posts, each defect mapped to the specific FTC Section 5 or state-law requirement it violates, plus a per-post disclosure template the agency must use going forward.
See a reference example
Duolingo, mascot-persona disclosure audit (excerpt) POST 1: 'Duo here, obsessed with this new streak feature.' DEFECT: none found, caption includes '#DuolingoPartner AI-generated persona' per post VERDICT: compliant POST 2: Story reshare with sponsorship tag only DEFECT: missing synthetic-persona disclosure, sponsorship tag alone is insufficient VERDICT: revise before publishing
Success criteria
You're done when you can:
- Flags all three drafts as non-compliant with a specific rule cited for each
- Correctly explains why a sponsorship-only tag or a bio-only disclosure fails the per-post standard
- Does not flag stylistic choices (hashtag placement, phrasing) as compliance defects