Acceptable Use & Prohibited Content Policy

This Acceptable Use & Prohibited Content Policy (the "Policy") governs all use of https://velum.ai and the Velum AI platform, portal, tools and APIs (collectively, the "Service"), operated by Velum AI operator [to be configured] ("Velum AI", "we", "us"). It applies to every account holder, creator, agency, team member, and anyone who uploads material to, generates content through, or otherwise accesses the Service ("you").

The Service exists for a single, narrow purpose: verified adult (18+) content creators upload photographs of themselves so that we can train a per-creator artificial-intelligence likeness model (a LoRA or "digital clone") of that same consenting creator, and generate new adult images and video of the creator's own verified likeness, which the creator may then sell on third-party adult platforms. Any use outside that purpose is prohibited.

This Policy is incorporated into and forms part of our Terms of Service, and must be read together with our Privacy & Biometric Data Policy, our 18 U.S.C. § 2257 Compliance Statement, and our Consent & Content-Generation Policy. Where this Policy is stricter than another document, the stricter rule controls. Capitalized terms not defined here have the meaning given in the Terms of Service.

This Policy is enforced against the same standard everywhere we operate, regardless of the location of the creator or the viewer. In some jurisdictions our rules are stricter than local law; that is intentional. Nothing here is legal advice to you.

You may create, train and generate content using one likeness only: your own. Before any model is trained we verify your identity and age using a government-issued photo ID and a liveness/verification selfie, and we bind your account to that verified identity. Every training image you upload must depict you, the same verified person, and no one else as a subject.

You may not upload, request, train on, or generate content that depicts any other real, identifiable person — whether a celebrity, an ex-partner, a friend, a co-worker, a stranger, another creator, a composite "face-swap", or a fictional-seeming likeness that is in fact based on a real person. Requesting a clone of anyone other than yourself is a serious violation and, depending on the content, may also be a crime.

Group, duet or "scene-partner" content is permitted only where every human likeness that appears is either (a) your own verified likeness, or (b) a separately verified, consenting adult creator who has an active Velum AI account and has signed a scene-specific consent for that exact collaboration, documented through the Service. Fully synthetic non-real "partners" are subject to the same age, non-minor and thematic rules in Sections 5–9. You are responsible for ensuring no real third party is recognizably reproduced in the background, on screens, in reflections, or in any incidental element of a generated image.

Adding text, watermarks, captions or metadata that misattribute the content to another person, or that suggest a real third party participated when they did not, is prohibited.

We treat the creation of content and the publication of content as two separate acts, each requiring its own basis. Your signed onboarding consent authorizes us to build and operate your digital clone and to generate content for you. It does not, by itself, authorize the public distribution of any particular image or video.

You are solely responsible for where, whether and how you publish or sell the content we deliver to you, and for complying with the terms, age rules, labeling rules and prohibited-content rules of every third-party platform on which you publish. Content that is acceptable to generate for private review may still be unacceptable to publish on a given platform.

You may withdraw or narrow your consent at any time as described in our Consent & Content-Generation Policy and Privacy & Biometric Data Policy. Withdrawal stops future generation and triggers deletion of your training data and model on the timelines stated in those policies, but cannot retroactively recall content you have already downloaded, published or sold.

Because we generate the content, Velum AI (Velum AI operator [to be configured]) acts as the producer of record for the images and video created through the Service and maintains age-verification and consent records for the verified individual depicted, as described in our 18 U.S.C. § 2257 Compliance Statement. Records custodian and inspection address: «[TO BE SET: Custodian of Records legal name and U.S. street address]».

Every generation request is subject to review before content is finalized and released to you for download. Review combines automated filtering (including prompt screening, nudity/age classifiers, and hash-matching against known-illegal-content databases) with human moderation of flagged items. We may hold, refuse, edit-block, or delete any request or output that violates this Policy, and we may do so before, during or after generation.

Passing review is not an endorsement, legal clearance, or guarantee that content complies with the rules of any third-party platform, or with the law of any particular jurisdiction. Review is a safety and compliance control operated for Velum AI's purposes; the ultimate responsibility for lawful, in-policy requests remains yours.

We may re-review previously generated content at any time, including retroactively when detection tools, card-network rules, or legal obligations change.

We have absolute zero tolerance for any content that sexualizes, or appears to sexualize, a minor. This is the single most important rule on the platform and it admits no exception, artistic claim, role-play framing, or "it's only AI" defense.

The following are strictly and permanently prohibited: (a) any sexual or nude content depicting a real person under 18; (b) any AI-generated, drawn, rendered, morphed or "virtual" content that depicts, or is designed to appear to depict, a minor in a sexual or nude context, including fully synthetic characters; (c) age-regression or "age-down" generations that make your own verified adult likeness appear to be a minor; (d) content styled to suggest a minor — for example childlike bodies, school or nursery settings, toys, pacifiers, braces, or "barely legal"/"teen" framing that depicts or implies a person under 18; and (e) any prompt, caption, tag, or metadata that solicits, describes or markets any of the foregoing.

Under U.S. law such material is child sexual abuse material (CSAM), and depending on content and jurisdiction may also constitute obscene visual representations of the sexual abuse of children. We do not require that a request "succeed" to act: attempting to generate this content is itself a terminating violation.

When we detect apparent CSAM or attempts to produce it, we will: immediately block and preserve the material and associated account data as required by 18 U.S.C. § 2258A and applicable law; report to the National Center for Missing & Exploited Children (NCMEC) through the CyberTipline; permanently terminate the account and the person's access; and cooperate with law enforcement, including preserving evidence for the statutory retention period. We do not provide advance notice of a report, and we do not "warn and allow" for this category.

All content on the Service must depict only the fully informed, freely given consent of the verified person(s) whose likeness appears. Content that depicts, sexualizes, or is designed to appear to depict non-consent is prohibited, and content that reproduces a real person without their consent is prohibited.

Prohibited non-consensual content includes, without limitation: (a) any depiction of rape, sexual assault, "forced" sex, or coercion presented as real or as the theme of the scene; (b) content depicting a person who is asleep, unconscious, drugged, heavily intoxicated, hypnotized, or otherwise incapacitated and therefore unable to consent; (c) "revenge porn", leaked, hacked, hidden-camera, upskirt, voyeur, or otherwise surreptitiously obtained material; (d) sextortion or any content produced under threat, blackmail or duress; and (e) any deepfake, face-swap, or likeness-based content that reproduces a real, identifiable person who has not personally verified and consented through the Service (see Section 2).

You may not upload another person's private images to be cloned, morphed, "nudified", or inserted into generated content, even if you claim to have their permission — the only permission we accept is that person's own verified, documented consent given directly to Velum AI.

Consensual depictions of fantasy dynamics between verified adults may be permissible only where they are clearly consensual within the scene and are not otherwise prohibited by Sections 5, 8 or 9 or by our payment partners' rules; where a theme is ambiguous, we will treat it as prohibited.

You may not impersonate any person or entity, or generate content that falsely suggests a real person's participation, endorsement, or authorship. This includes generating content "in the style of", "as", or resembling a named celebrity, public figure, influencer, or private individual.

You may not upload reference images, prompts, or textual descriptions intended to steer a generation toward the likeness, distinctive features, tattoos, or identifying marks of any real person other than yourself (or a co-verified scene partner under Section 2).

You are responsible for third parties who appear incidentally in your uploaded photographs (for example, people visible in a mirror, a window, a photo-within-a-photo, or the background). Remove or fully obscure them before upload; we may reject uploads that contain unverified faces.

Trademarks, logos, uniforms (including those implying law-enforcement, military, or medical authority in a deceptive way), and other brand or institutional identifiers may not be used in a manner that infringes rights or misleads viewers.

The following themes and depictions are prohibited on the Service, both as generated output and as the subject of prompts, tags, titles or marketing, because they are illegal in relevant jurisdictions, prohibited by our card networks and payment processors, or both. This list is illustrative, not exhaustive; we may prohibit additional categories to remain compliant with applicable law and our payment partners.

(a) Incest and familial content — any depiction, description or theme of sexual activity between blood relatives, including "pseudo-incest" or step-family framing presented as a family relationship. (b) Bestiality / zoophilia — any sexual content involving animals or animal-like creatures. (c) Necrophilia and "snuff" — any sexualization of death, corpses, or content depicting or simulating a person being killed. (d) Minor-adjacent or age-play content of any kind (see Section 5, which controls absolutely).

(e) Non-consent themes as described in Section 6, including drugged, unconscious, hypnosis, and "forced" scenarios. (f) Extreme violence, torture, mutilation, dismemberment, serious or permanent bodily injury, cannibalism, or content whose primary purpose is to depict pain or gore. (g) Weapons used to threaten, coerce, or inflict non-consensual harm in a sexual context. (h) Breath-play, strangulation, or asphyxiation depicted as non-consensual or life-endangering.

(i) Scat/coprophilia, and other bodily-waste content to the extent restricted by our payment processors. (j) Content depicting or promoting self-harm, suicide, or eating disorders in a sexualized context. (k) Any hate-based sexual content that dehumanizes a person or group on the basis of a protected characteristic. Because payment-processor rules vary and are updated frequently, certain fetish categories (for example, some watersports, heavy BDSM, or fluid content) may be conditionally restricted; the current governing list is «[TO BE SET: current payment processor(s) and the version/date of their published prohibited-content list]».

You may not use the Service to create, request, promote, or facilitate any content that is illegal, or that depicts, advertises, or solicits illegal conduct. This includes, without limitation: human trafficking, sexual exploitation, or any content connected to trafficking or coercion; prostitution, escort solicitation, or the sale of in-person sexual services; the sale, use, or promotion of illegal drugs; and any activity that violates applicable obscenity, exploitation, or sexual-conduct laws.

You may not use the Service to harass, threaten, stalk, dox, defame, or endanger any person, or to produce content intended to intimidate or extort.

You may not use generated content for fraud, catfishing, romance scams, fake dating or fan profiles, synthetic identity creation, or to deceive any person about who they are interacting with.

You may not upload malware, attempt to access accounts or data that are not yours, scrape or reverse-engineer the models or the Service, circumvent moderation or safety controls, resell or sublicense access to the Service without a written agreement, or use the Service to build, train, or benchmark a competing likeness model.

The Service and the content generated through it must at all times comply with the rules of the card networks and payment processors we use to accept payment, including Visa (including its Global Brand Protection Program and standards for merchants facilitating adult content), Mastercard (including its Specialty Merchant Registration Program requirements for adult content), and our adult-billing payment processor(s), which may include Segpay, CCBill, Epoch, or another authorized processor.

In furtherance of those rules, and consistent with the rest of this Policy: (a) every person depicted is age- and identity-verified and has documented, revocable consent; (b) all content is reviewed before it is released; (c) illegal content — above all CSAM and non-consensual content — is absolutely prohibited, blocked, and reported; (d) we maintain a monitored complaint and takedown process (Sections 16–17); and (e) we keep the records the networks and processors require, and we cooperate with their audits and requests.

If a card network or payment processor updates its prohibited-content standards, those updated standards apply to you automatically upon our adoption of them, and may narrow what is permitted under this Policy without further notice. Complaints subject to processor rules will be acknowledged and resolved within «[TO BE SET: complaint acknowledgment/resolution timeframe required by current processor(s), e.g., 7 business days]».

Content the card networks require to be prohibited will be prohibited on the Service even if it would otherwise be lawful in a given jurisdiction.

All content produced through the Service is synthetic media created by artificial intelligence from your verified likeness. You may not represent generated content as an unaltered photograph or as documentary of an event that did not occur where doing so would deceive viewers, defraud them, or violate the disclosure rules of a platform or applicable law.

We may embed and preserve provenance signals in delivered content — for example, content credentials/metadata (such as C2PA-style manifests), invisible or visible watermarks, or model tags — to support authenticity and abuse detection. You may not strip, alter, or falsify these provenance signals, and you must not misrepresent third-party content as having been produced by Velum AI.

Where a third-party platform, advertising rule, or law (including emerging AI-transparency and synthetic-media disclosure requirements) requires that AI-generated or AI-altered sexual content be labeled as such, you are responsible for applying that label when you publish. We recommend clear "AI-generated" disclosure as a default best practice.

Beyond content, the following conduct is prohibited: creating more than one account to evade verification, suspension, or limits; sharing, selling, or transferring your account or your trained model to another person; submitting false verification documents or another person's identity documents; and using automated means to overload, probe, or disrupt the Service.

You may not attempt to jailbreak, prompt-inject, or otherwise manipulate our moderation, safety, or age-classification systems, or to coax the models into producing content prohibited by Sections 5–9. Attempts to do so are themselves violations, independent of whether they succeed.

You may not use the Service in a way that infringes intellectual-property rights, misappropriates another's name, image or likeness (NIL) or right of publicity, or breaches any contract or platform term to which you are subject.

We use a combination of automated and human systems to enforce this Policy, including prompt and caption screening, image classifiers for nudity and apparent age, and hash-matching of uploaded and generated media against industry databases of known illegal content. Flagged material is routed to trained human reviewers.

We may log prompts, uploads, outputs, and account activity as needed to operate these controls, investigate suspected violations, meet record-keeping obligations, and respond to lawful requests. Handling of this data — including biometric data — is described in our Privacy & Biometric Data Policy.

Automated detection is imperfect and may over- or under-flag; human review, account action, and mandatory reporting are layered on top precisely because no single control is sufficient for this content category.

We enforce this Policy proportionately to the severity of the violation, but we reserve the right to take the most serious action for any violation. Consequences may include, alone or in combination: refusing or silently blocking a request; removing or quarantining generated content; issuing a warning; restricting features; suspending the account pending investigation; and permanently terminating the account and deleting the associated model and data.

For the zero-tolerance and clearly illegal categories in Sections 5, 6 and 9 — in particular any apparent CSAM, non-consensual, or trafficking-related content — the response is immediate and permanent termination, preservation of evidence, and reporting to authorities, without warning and without any right to reinstatement.

Termination for a violation does not entitle you to a refund, and does not limit any other remedy available to us. We may decline to do business with, and may permanently ban, any person we have terminated for a serious violation, including across future accounts. Where a violation causes us loss (for example chargebacks, fines from card networks, or legal costs), you remain responsible for it to the extent permitted by law.

We are legally required, and we are committed, to report apparent child sexual abuse material to the National Center for Missing & Exploited Children (NCMEC) via the CyberTipline, and to preserve the associated content and records as required by 18 U.S.C. § 2258A. We will report and cooperate regardless of where the user or the content originated.

We may also report other suspected crimes — including non-consensual intimate imagery, trafficking, sextortion, and threats of violence — to appropriate law-enforcement authorities, and we may respond to valid legal process such as subpoenas, warrants, and preservation requests.

When we make a mandatory or law-enforcement report, we may preserve and disclose relevant account, verification, prompt, upload, output, and log data. We generally will not tip off a user that they are the subject of such a report where doing so is prohibited by law or would compromise an investigation or child-safety.

We operate an expedited removal process for non-consensual intimate imagery (NCII). Consistent with the TAKE IT DOWN Act, if you are an identifiable individual depicted in intimate visual content published or hosted by the Service without your consent (including a digital forgery of you), you may request removal, and we will remove the reported content, and make reasonable efforts to remove identical copies, within 48 hours of a valid request. Send NCII removal requests to «[TO BE SET: NCII/removal intake email or web form]».

Because our model requires your own verified consent before any likeness is cloned, non-consensual likeness content should never exist on the Service; this process exists as a backstop and for reports concerning content we host or generate.

We respond to notices of alleged copyright infringement under the U.S. Digital Millennium Copyright Act (DMCA). If you believe content on the Service infringes your copyright, send a compliant notice to our designated agent: «[TO BE SET: DMCA designated agent name, mailing address, email, and U.S. Copyright Office registration number]». We will remove or disable access to infringing material and terminate repeat infringers. Counter-notice procedures are available as provided by law.

These removal channels are in addition to, and do not replace, the reporting channel in Section 17.

If you become aware of content or conduct that violates this Policy — including content depicting you or another real person without consent, suspected exploitation of a minor, impersonation, or any prohibited category above — report it to us immediately at Support contact pending configuration, or through the in-portal report control where available.

To help us act quickly, please include (where you can, and without sending illegal material yourself): a description of the content or conduct, any URL, account name, or content identifier, why you believe it violates this Policy, and how we can contact you. If a minor is in immediate danger, contact local emergency services first.

We review reports promptly, prioritize child-safety and non-consensual-content reports above all others, and take action under Sections 14–16 as appropriate. We aim to acknowledge reports quickly and to resolve them within the timeframe required by our payment partners and applicable law. We prohibit retaliation against anyone who makes a good-faith report.

To verify your identity and train your model, we process biometric identifiers and biometric information derived from your face and body ("biometric data"), which is sensitive personal data under laws such as the Illinois Biometric Information Privacy Act (BIPA) and special-category data under GDPR Article 9. We collect and use it only with your explicit written consent, only to operate the Service for you, and we do not sell it.

We retain training images, the trained model, and verification records only for as long as needed to provide the Service and to meet our legal and record-keeping obligations (including § 2257 and card-network requirements), and we destroy biometric data on the retention schedule set out in our Privacy & Biometric Data Policy, or sooner upon a valid deletion request, subject to legal holds.

Depending on where you live, you may have rights to access, correct, delete, or restrict the processing of your personal data, to withdraw consent, and to lodge a complaint with a regulator (for example, GDPR/UK GDPR rights, and CCPA/CPRA rights). Exercise these rights, and see full detail, through our Privacy & Biometric Data Policy. Our EU representative under GDPR Article 27 is «[TO BE SET: EU Article 27 representative name and address]» and our UK representative is «[TO BE SET: UK GDPR representative name and address]».

We may update this Policy from time to time, including to reflect changes in law, card-network or payment-processor rules, detection technology, or the Service. Material changes will be posted with a new "last updated" date, and continued use of the Service after an update constitutes acceptance of the revised Policy. Changes required for legal or payment-compliance reasons may take effect immediately.

This Policy is part of, and is governed by the same terms as, our Terms of Service, including their governing-law and dispute-resolution provisions. Governing law and venue: «[TO BE SET: governing-law state and venue]».

Questions about this Policy, and reports of prohibited content, may be directed to Velum AI operator [to be configured] at Support contact pending configuration. Records-inspection, DMCA, NCII, and data-rights requests should use the specific contacts identified in Sections 4, 16 and 18. This document is an internal draft prepared for review by the operator's attorney and is not legal advice to any end user.