AI models / PRACTICAL CHECK
How should you introduce a fictional AI campaign model?
Introduce the person as a fictional AI campaign character and approve the surrounding role claims for each placement. A name and recurring face do not establish employment, qualifications or personal product use. Keep the exact caption with the asset version, then check the destination’s own disclosure process.
Choose the public role before writing a biography
Imagine an illustrative outdoor-accessory campaign using an adult fictional model named Rowan. The team wants to introduce Rowan on the brand's social page. Calling the character a fictional campaign model describes the creative role; calling Rowan a local guide who tests every bag creates factual claims about a person and experience that did not occur.
Write one accurate sentence explaining the role. Keep internal casting notes separate from public identity copy. A recurring face, name or wardrobe is an editorial choice, not proof of age, birthplace, qualifications or an employment relationship. Do not invent those details merely because a profile feels incomplete without them.
Decision first: if the task is an introduction, write the role without a life story. If it is an apparent staff profile, expert answer or customer account, ask for the real relationship and evidence before using that format. If the evidence does not exist, change the format as well as the wording. For actual recommendations, use the separate testimonial-authenticity check.
Review nouns as carefully as recommendations
Read proposed labels such as ambassador, designer, customer, employee and expert. Each can imply a relationship or qualification even when the character never says it recommends the product. Ask what an ordinary viewer would infer from the complete post, including account name, image caption and any first-person wording.
If a real person is involved, use only the identity and role information that is verified and authorized for that context. Do not use a fictional label to disguise a real-person reference or imply that permission issues disappear. Keep adult personal references within the intended workflow and have unresolved identity questions reviewed before publication.
Read the introduction with the destination heading. “Meet Rowan” beneath “Our campaign characters” differs from the same sentence beneath “Meet our designers.” Review profile fields, image alt text, badges and linked landing pages as part of the message. Removing “I recommend” does not correct a false job title elsewhere.
Use an identity-copy matrix
| Public description | What it implies | Evidence or revision needed |
|---|---|---|
| Fictional AI campaign character | A deliberately created visual role | Clear generated-character context |
| Real model named in the caption | An identifiable person participated | Verified identity and appropriate permission |
| Our product designer | An employment or creative contribution | Actual relationship and contribution records |
| Independent customer | Real purchase or use outside the brand | Authentic first-hand evidence |
| Outdoor guide with years of experience | A biography and professional history | Use a fictional-role description if those facts do not exist |
Keep the distinction understandable where people encounter the content. A disclosure in an unrelated footer may not explain a realistic profile image and biography. Prefer ordinary wording near the introduction, and check the platform's own requirements for the destination.
Prepare the model and its disclosure separately
Select and review the visual reference in Models. This saves a reference for creation; it does not approve a public name, profession or biography. Keep the chosen reference version with the model handoff record so a later caption is attached to the right character.
YouTube’s Disclosing use of GenAI content guidance, checked October 21, 2026, requires disclosure for realistic AI-generated or meaningfully altered content, with specified exceptions for non-realistic content and minor edits. The current instructions use an AI use setting during upload. This is a YouTube-specific process, not a universal legal rule or a substitute for accurate campaign copy.
For an image used only on a product page, record the page’s intended role and applicable destination requirements. For a realistic video uploaded to YouTube, separately record the upload setting and visible final context. A platform label cannot turn an invented professional history into a verified biography.
Approve reuse of the description with each asset
Fictional worked example, not a completed campaign test: the outdoor brand introduces Rowan with “Rowan is our fictional AI campaign character, used to illustrate accessory styling.” That explains the role without claiming a real trip or purchase. A proposal saying “Rowan, our field tester, carried this bag for a year” is held because no person, employment or use evidence supports it.
A second publisher reuses the approved picture on a staff page. The previous caption approval does not cover that new implication. Keep it out of the staff directory or obtain a different, truthful presentation. A third use adds a speech bubble describing a purchase; route that to the testimonial review rather than approving it as a harmless caption refresh.
After the copy owner approves the exact text and destination, inspect the final rendered post. Confirm that a crop, abbreviated bio or repost did not remove the fictional-role context. Save the actual URL or placement preview and unresolved observations. Do not mark “disclosed” merely because the draft document contains a label. Use the image review checklist separately for face, hands and product accuracy.
Public identity record for one actual placement
Copy this external record for one actual asset and destination. Fill it from real observations; leave unknown items pending. The empty cells are not completed checks.
| Record | Actual observation or decision |
|---|---|
| Character and saved reference version | |
| Asset filename or identifier | |
| Destination, page heading and proposed public role | |
| Exact approved name, biography and caption wording | |
| Evidence for any real relationship, or claim removed | |
| Applicable destination disclosure decision and setting | |
| Actual final context observed, date and reviewer | |
| Pending issue, reuse restriction and release decision |
Sources and further reading
Recorded source-review dates are listed individually below. An undated source does not inherit the article’s publication date. Software and platform rules can change; check the current original before acting.
- CartVisual: product scope and editorial policy · Reviewed
- YouTube: Disclosing use of GenAI content · Reviewed
Frequently asked questions
Can a fictional AI model have a name?
Yes, when the presentation makes the fictional role clear and does not falsely identify a real person or imply an unsupported biography.
Is fictional character wording enough on every platform?
No. Check destination-specific synthetic-content and advertising disclosure rules as well as the full image and copy context.
Can I describe the character as a customer without a testimonial?
That still implies a real customer relationship. Use an accurate fictional campaign-role description when no authentic customer evidence exists.
Does an approved caption cover a staff page or repost?
No. Recheck the actual heading, profile fields, caption and destination. A new placement can imply employment or experience that the approved fictional-character introduction did not claim.
PREPARE ASSETS FOR YOUR NEXT STEP
Create supporting product assets in CartVisual.
Create new product creative assets in the workspace, then continue in your chosen tools. Third-party software procedures described in this guide take place in those applications; this workspace link does not imply an integration. Check current capabilities, account access and credit quotes.
Explore asset creation tools Check pricing and credits