AI Content Disclosure Rules for OnlyFans Agencies

Last updated: September 15, 2026

The Disclosure Obligation In One Extractable Block

The AI content disclosure rules for OnlyFans agency generated content operate at the post level, not the account level. Every piece of content an agency publishes, including enhanced photos, fully synthetic scenes, face swaps, or AI chatters, maps to a specific disclosure action. OnlyFans’ Acceptable Use Policy (last updated September 2025) states: “AI Generated content must comply with our Terms of Service and must be clearly and conspicuously captioned as AI Generated Content with a signifier such as #ai, or #AIGenerated.” A bio disclosure alone does not satisfy this requirement.

Accepted disclosure strings, in order of strength:

• Caption Line: “This image was created with AI assistance”
• Hashtag: #AI or #AIGenerated
• Bio Disclosure: “This profile features AI-generated content”
• Pinned-Post Disclosure: “All AI-generated content on this profile is labeled #AIGenerated”

With the core obligation established, agencies can anchor the rest of their workflow on these rules.

The Disclosure Decision Tree: Rules For Each Type Of Post

Apply this decision tree to every asset before it leaves the queue. The key issue is the type of AI involvement and the matching disclosure that specific use requires.

1. Unmodified Photo

No AI disclosure is required. Standard content rules apply. Proceed to publication.

2. AI-Enhanced Or Upscaled Photo

Disclosure is required only when AI materially changes the image. Routine post-production like color grading, skin smoothing, and upscaling does not trigger the requirement. If the AI altered a body part, generated a background, or created a scene that never happened on camera, label it. When in doubt, label.

This category sits between simple editing and full generation, so treat any creative alteration as AI content.

3. AI-Generated Scene Using A Real Creator’s Likeness

Use a mandatory #AI or #AIGenerated caption. AI-generated content must clearly resemble and trace back to the identity-verified account holder. AI may modify or enhance the real creator, such as touch-ups, digital makeup, and lighting, and may generate backgrounds or scenes, but the output must feature the verified account holder. Confirm creator resemblance before publishing. This rule protects against accidental deepfakes of non-consenting people.

4. AI Face Swap

Use a mandatory #AI or #AIGenerated caption on every post. Face-swapping a creator’s face onto unverified third-party adult performers is banned. Face-swapping public figures or acquaintances into explicit scenes violates OnlyFans’ policy, which prohibits deepfakes and AI content depicting anyone other than the verified creator, and creators who use AI are barred from making money on the site. Both face-swapping onto unverified third parties and face-swapping public figures fall into the prohibited deepfake category.

5. Fully Synthetic Character With No Source Person

Use a mandatory #AI or #AIGenerated caption on every post. This content type is permitted only if the operator completes standard identity verification, every post carries the #ai signifier, and the persona’s face is original rather than generated from a real person’s likeness. Add a pinned-post disclosure and bio disclosure in addition to per-post captions. This combination shows the audience and the platform that the character is fictional.

6. AI Chatter Or Automated DM

Treat AI chatters as a separate compliance issue governed by state law and platform policy on misleading information. See the AI Chatters section below for the detailed standard.

See how Sozee automates these disclosure steps across your roster.

Exact Disclosure Language: Caption, Bio, And Pinned-Post Strings

A caption-level disclosure is stronger and more portable than relying on a profile bio that may not appear beside every post, message, preview, or shared screenshot. Place the tag at the start of the caption, not buried in a hashtag block. Use the following strings as copy-paste templates.

Caption Strings:

• “This image was created with AI assistance. #AI”
• “Pool day with [character name]. Fictional AI-generated character. #AIGenerated”

Bio Disclosure Strings:

• “This profile features 100% synthetic, AI-generated content.”
• “All content on this profile is AI-generated. #AIGenerated”

Pinned-Post Disclosure String:

• “All AI-generated content on this profile is labeled #AIGenerated. This profile features a fictional AI character. No real people were used, harmed, or impersonated. This content is for adults only.”

OnlyFans Platform Rules, FTC Rules, And State Likeness Law

Three separate regimes apply simultaneously to every post. Satisfying one regime does not satisfy the others. The table below maps each regime to its operative requirement.

OnlyFans Platform Rules are enforced fastest. OnlyFans’ enforcement escalation ladder starts with administrative warnings and content removal for first-time labeling offenses. Repeated violations lead to 7- to 30-day suspensions. Deepfakes or unverified synthetic accounts result in immediate permanent termination and forfeiture of funds.

FTC Rules apply to the agency as the publishing intermediary. 16 CFR 255.1(f) states that advertising agencies, public relations firms, and other similar intermediaries may be liable for their roles in creating or disseminating endorsements containing representations they know or should know are deceptive. The FTC’s July 2026 AI policy statement (File No. P264200, 91 FR 41638) reaffirms that AI-related practices remain subject to the FTC’s general deception and unfairness authority under Section 5.

State Likeness Law governs consent before generation as well as disclosure after publication. California AB 2602, effective January 1, 2025, bars a contract clause from authorizing a digital replica of a living performer’s voice or likeness without informed, specific consent negotiated with proper representation. Tennessee’s ELVIS Act (Tenn. Code Ann. § 47-25-1101 et seq., effective July 1, 2024) extends right-of-publicity protection to AI simulations of a person’s voice. New York General Business Law § 396-b(3) requires conspicuous disclosure when a commercial advertisement uses a “synthetic performer.”

ChatGPT has stated it cannot verify a current official OnlyFans rule requiring a standardized AI-generated label. This article resolves that ambiguity. The requirement appears in OnlyFans’ Acceptable Use Policy, last updated September 2025, under the heading “AI Generated Content.” The comparison query “FTC AI disclosure rules vs. OnlyFans platform rules” has a direct answer: both apply, and the platform rule is enforced faster.

AI Chatters And Automated DMs: Disclosure For Conversations

OnlyFans AI chatter disclosure is a distinct compliance issue from AI media disclosure. Treat conversational AI as its own risk category.

OnlyFans’ Acceptable Use Policy contains no mention of “AI chatbot,” “chat bot,” or “automated messaging.” The closest AI-related restriction is item 9, which prohibits “Any explicit image of another person without their consent including an artificially generated image” and focuses on non-consensual imagery, not chat disclosure. However, OnlyFans’ Terms of Use define a “Creator Interaction” to include “any other interaction or payment between a User and a Creator’s account or Content, including direct messages,” placing DMs expressly within the scope of creator responsibility. Acceptable Use Policy item 8 forbids “Inaccurate information including misleading descriptions of media or account information.” An autonomous bot impersonating the creator in DMs counts as a misleading description of account information.

State law is moving faster than platform policy on this question. California’s SB 243, effective January 1, 2026, established the first mandatory AI chatbot disclosure requirements for consumer-facing interactive AI in the United States. Oregon’s SB 1546 and Washington’s HB 2225 were both enacted in March 2026 and take effect January 1, 2027. They require operators of AI companion chatbots to clearly disclose that the user is interacting with an artificially generated system, not a natural person. Washington’s HB 2225 requires the alert to be redisplayed every three hours during lengthy conversations.

Use this operational guidance. Use AI to assist a trained human who reviews and sends every message, keep a defined persona voice, and give the creator transcripts on request. That approach captures AI’s speed while avoiding the deception liability that comes with letting a bot speak as the creator.

Agency Documentation And Consent: Logs And Contract Language

An audit trail is mandatory for serious agencies. Payment processors now require model provenance and consent records before agreeing to process AI-driven adult content. Build the ledger before a platform or processor asks for it.

What To Log For Every Asset:

• Prompt text
• AI tool name and version
• Generation timestamp
• Performer consent forms
• Model metadata: model name, version, prompt templates, LoRA or checkpoint identifiers
• Output hashes
• Signed attestations

What A Consent Clause Must Contain:

• Source authorization, covering whether the uploader may possess and submit the source file
• Transformation consent, covering whether the subject agreed to this specific kind of change
• Distribution consent, covering whether the subject agreed to the result being posted publicly or used commercially
• Revocation and remedy terms, covering how the subject can withdraw permission, report abuse, and request removal

Agency-Side Documentation Checklist:

• Verified account status: government ID plus active biometric liveness check
• Confirmed creator resemblance in every AI photo or video
• Audit for third-party or celebrity likenesses
• Explicit #AI or #AIGenerated labeling in the caption
• Audit-ready storage of 2257 age records and signed AI likeness agreements
• Encrypted compliance ledger with model consent releases, synthetic provenance logs, and dated disclosure captions

Enforcement And Consequences: What Happens When Disclosure Is Missing

OnlyFans’ Acceptable Use Policy states: “Any breach of this Acceptable Use Policy may result in deactivation of your Content and/or your account.” As noted earlier, OnlyFans’ enforcement ladder escalates from warnings and content removal to suspensions and then permanent termination with forfeiture of funds for deepfakes or unverified synthetic accounts.

The agency is a liable party, not only the creator. 16 CFR 255.1(f) states that advertising agencies and other similar intermediaries may be liable for their roles in creating or disseminating endorsements containing representations they know or should know are deceptive. The federal TAKE IT DOWN Act (Public Law 119-12, signed May 19, 2025) establishes criminal liability for disseminating non-consensual intimate imagery including synthetic media, with platforms required to remove flagged content within 48 hours.

The One-Page SOP: Hand This To Your VA

This SOP turns the rules above into a repeatable checklist your VA can follow for every post.

Before Publishing

• Verify the account holder is a verified human creator with government ID and active biometric liveness check.
• Confirm the AI output features the verified account holder’s real face and body.
• Audit for third-party or celebrity likenesses.
• Confirm the character is clearly adult (18+).

Caption Requirements

• Place #AI or #AIGenerated at the start of the caption.
• Add a caption line: “This image was created with AI assistance.”
• For fully synthetic characters, add: “Fictional AI-generated character.”

Bio Requirements

• Add to bio: “This profile features AI-generated content.”

Pinned-Post Requirements

• Pin a post stating: “All AI-generated content on this profile is labeled #AIGenerated.”

Documentation Steps

• Log prompt, tool version, timestamp, and performer consent.
• Store all records in an encrypted compliance ledger.
• Retain documentation for at least two years.

Put this SOP into action with Sozee’s compliance-ready workflow.

Sozee As A Practical Tool For This Compliance Playbook

The AI content disclosure rules in this playbook require per-post discipline, per-character consistency, and an audit trail that survives a platform review. Sozee helps agencies execute all three without adding headcount.

Sozee’s locked likeness keeps every asset in a deliverable looking like the same person on the same day, which supports the Creator Resemblance Rule. Reusable environments, outfits, and objects compound across shoots so agencies build a character’s world once and publish from it repeatedly. Photo Control’s five directable dimensions, covering Setting, Outfit, Shot style, Expression, and Object, replace prompt gambling with deliberate direction. Photo Shoot turns one image into a locked, coherent set of up to ten, including a full SFW-to-NSFW arc paced by the operator.

The Scheduler connects Instagram, TikTok, X, Facebook, Reddit, and Fanvue per character, with a caption per platform. The #AIGenerated tag goes into the caption template once and publishes correctly every time. Analytics separate what Sozee posted from what the operator posted, creating a timestamped record of what was published and when. Teams and isolated workspaces let an agency run its entire roster from one login, with each character’s compliance documentation kept separate. Compliance and verification live in setup rather than being bolted on afterward. Models are private, isolated, and never used to train anything else.

The documentation, per-character publishing, and audit trail this playbook prescribes match Sozee’s default workflow, so compliance becomes part of everyday production.

Explore Sozee and align your production workflow with this compliance playbook.

Frequently Asked Questions

Do AI-Generated OnlyFans Photos Need A Hashtag Or A Caption Disclosure?

OnlyFans’ Acceptable Use Policy (last updated September 2025) requires AI-generated content to be “clearly and conspicuously captioned as AI Generated Content with a signifier such as #ai, or #AIGenerated.” Use both a hashtag and a caption line when possible. As noted earlier, caption disclosures are more portable than bio disclosures. Place the tag at the start of the caption rather than buried in a hashtag block. A bio disclosure and a pinned-post disclosure add redundancy but do not substitute for the per-post caption requirement.

What Disclosure Is Required For AI Face Swaps Vs. AI Enhancement?

AI face swaps require a mandatory #AI or #AIGenerated caption on every post. Face-swapping a creator’s face onto unverified third-party adult performers is banned outright under OnlyFans’ rules. Face-swapping public figures or acquaintances into explicit scenes violates OnlyFans’ policy, which prohibits deepfakes and AI content depicting anyone other than the verified creator, and creators who use AI are barred from making money on the site. AI enhancement requires disclosure only if AI materially changed the image. Routine post-production like color grading, skin smoothing, and upscaling does not fall under the labeling clause. If the AI altered a body part, generated a background, or created a scene that never occurred on camera, label it. When the line is unclear, label it.

Do AI Chatters And Automated DMs Need To Be Disclosed?

OnlyFans’ Acceptable Use Policy contains no mention of “AI chatbot,” “chat bot,” or “automated messaging.” As discussed in the AI Chatters section, OnlyFans’ Terms of Use include DMs in Creator Interactions, and AUP item 8 prohibits misleading descriptions. At the state level, California’s SB 243 (effective January 1, 2026), Oregon’s SB 1546, and Washington’s HB 2225 (both effective January 1, 2027) require AI companion chatbots to disclose that the user is interacting with an artificially generated system. The operational standard remains simple: use AI to assist a trained human who reviews and sends every message, and avoid any autonomous bot that speaks as the creator.

What Must An OnlyFans Agency Log For AI Content Compliance?

Log prompt text, AI tool name and version, generation timestamp, performer consent forms, model metadata such as model name, version, prompt templates, and LoRA or checkpoint identifiers, output hashes, and signed attestations. Store everything in an encrypted compliance ledger and retain it for at least two years. A consent clause must contain four distinct permissions: source authorization, transformation consent, distribution consent, and revocation or remedy terms that explain how the subject can withdraw permission, report abuse, and request removal or account action. Payment processors now require model provenance and consent records before agreeing to process AI-driven adult content, so the ledger supports both platform compliance and payment processing continuity.

What Is The Difference Between OnlyFans AI Content Policy 2026 And FTC AI Disclosure Rules?

Both regimes apply simultaneously to every post an agency publishes. OnlyFans’ platform rules are enforced fastest. Violations draw warnings, suspensions, or permanent termination depending on severity, and the platform acts before any regulator does. The FTC’s rules under 16 CFR Part 255 and Section 5 of the FTC Act apply to the agency as the publishing intermediary, with liability for creating or disseminating endorsements the agency knows or should know are deceptive. The FTC’s July 2026 AI policy statement (File No. P264200, 91 FR 41638) reaffirms that AI-related practices remain subject to this authority. State likeness laws, including California AB 2602, Tennessee’s ELVIS Act, and New York General Business Law § 396-b(3), govern consent before generation and disclosure in commercial advertising. Satisfying OnlyFans’ caption requirement does not satisfy FTC clear-and-conspicuous standards or state likeness consent requirements. All three regimes require separate, deliberate compliance actions.

Conclusion

The disclosure obligation operates per post, not per account. Every content type an agency publishes maps to a specific disclosure action, a specific consent record, and a specific line in the compliance ledger. The three regimes, covering OnlyFans platform rules, FTC authority, and state likeness law, apply simultaneously and remain non-interchangeable. AI chatter disclosure forms a separate compliance issue governed by state law that current platform policy does not fully resolve. The SOP above is the execution layer, and the documentation checklist is the audit defense. Sozee functions as the production platform that makes this workflow manageable at agency scale through locked likeness, per-character publishing, caption templates in the Scheduler, and compliance built into setup.

Start your free trial and see how Sozee simplifies AI disclosure and documentation.

Platform policy and FTC guidance last verified: September 15, 2026.