Acceptable Use Policy

Last updated: September 25, 2026

This policy is part of our Terms of Service. It says what you may create with PrintReady3D, how we screen content, how to report something, and how to appeal a decision.

Everything the AI generator produces is AI-generated, and we say so wherever it is shown.

What you can make

PrintReady3D exists to turn ideas into things you can print, so the bar is deliberately wide: figurines and miniatures, tabletop terrain, cosplay props and decorative replicas, household parts and organisers, planters, jewellery, toys, signs, gifts and models of your own designs or pets.

What is not allowed

A prompt or image is rejected — before anything is charged — and a model is removed after the fact when it falls into any of the following six categories.

• Pornography / NSFW. Explicit sexual acts, pornographic figures, and sexually suggestive depictions. Nudity in a classical or anatomical sense is judged by the surrounding prompt; the intent to titillate is the line.
• Violence / gore. Gratuitous gore, mutilation, and depictions that glorify real violence or real atrocities. Fantasy weapons on a figurine, monsters, and battle scenes for tabletop games are not what this rule is about, and they are allowed.
• Hate speech. Hate symbols and extremist insignia — for example Nazi iconography and terrorist group emblems — and content that attacks people for their race, ethnicity, religion, disability, sex, sexual orientation or gender identity.
• Child-unsafe content (CSAM). Any sexualised depiction of a minor, or of a figure that appears to be a minor, in any style or context. Zero tolerance, no exceptions, and we report it.
• Deepfakes / impersonation of real people. A real person's likeness used for harassment, impersonation, false endorsement, or to depict them in a way they have not agreed to. A figurine of yourself, or of someone who has agreed to it, is fine.
• Copyright / trademark infringement. Copyrighted characters, brand logos and protected designs reproduced as-is. Original designs inspired by a genre or style are fine; a copy of a specific character or a company's mark is not.

Absolutely prohibited

These are never allowed, in any context, and lead straight to the most severe enforcement below:

• Child sexual abuse material — listed twice on purpose. Confirmed cases are reported to the relevant authorities.
• Terrorism and violent extremism — recruitment material, glorification of attacks, or insignia of designated organisations.
• Weapons of mass destruction, and functional firearms or firearm parts — including printable guns, receivers, suppressors, and parts designed to make a weapon fire. Decorative props that cannot function are allowed.
• Non-consensual deepfakes — sexual or degrading depictions of a real, identifiable person.

Also not allowed

• Instructions or files for causing harm — anything whose value is the harmful function rather than the object, such as tools for breaking locks you do not own.
• Passing AI output off as human-made. Do not sell or present a model generated here as hand-sculpted or human-made original work when the buyer would care.
• Bypass attempts — wording designed to route around the screen, including obfuscation, misspellings, role-play framings and repeated attempts at the same rejected request.

How moderation works

1. A maintained blocklist, before anything else. Every text prompt is first checked against our own list of prohibited phrases — above all printable firearm parts, child-unsafe terms and extremist symbols, which general-purpose screens can miss when they are phrased as a 3D model. We add to it as new evasion attempts appear.
2. Automated prompt screening, before generation. The prompt is then checked by a third-party content-safety model — currently Meta Llama Guard 4, with OpenAI's moderation model as a fallback — before any credit is charged. The screen is fail-closed: if it is unreachable, the request is refused rather than allowed through. A rejected prompt costs nothing; rewriting the phrase that tripped the screen is usually enough.
3. Automated image screening, before use. Every image you upload is checked by an automated NSFW image classifier before it is used to make a model. A rejected image costs nothing.
4. Automated output screening, before you see it. Every concept image the AI generates or edits for you is checked by the same classifier before it is shown to you. A flagged result is withheld and anything charged for it is returned.
5. Human review of every report. Every public model page has a Report this model link, and a person reads every report we receive and decides what happens to the model.

We do not claim more than this. Automated screens are imperfect in both directions: they reject things they should not, and things slip through. That is what reporting and appeals are for.

Enforcement

• Level 1 — request rejected. The generation never runs and nothing is charged.
• Level 2 — model removed. We take the model down, refund the credits on request if you were charged for it, and email you which rule it broke.
• Level 3 — account closed. Repeated violations, or a single serious one, close the account; its remaining credits are forfeited.
• Level 4 — permanent ban and referral. Child sexual abuse material, terrorist content and the other absolutely prohibited categories end the account permanently and are referred to law enforcement where the law requires it.

The commercial-use licence on any model found to infringe someone else's copyright or trademark is revoked, and the model is removed.

Reporting a model

If you find a model on this site that breaks this policy — or that uses your likeness or your work without permission — use the Report this model link on its page, or write to us:

• Policy violations: support@printready3d.com
• Copyright and trademark: dmca@printready3d.com

Include the page URL and what is wrong with it. We reply within two business days and remove anything that violates this policy. We keep the reporter's identity confidential and never pass it to the person who made the model. Removal requests from the person depicted are honoured without argument.

Appeals

If a prompt or image was rejected, or a model was removed, and you think we got it wrong, email support@printready3d.com within 14 days, with the model link or the rejection message you were shown, and a sentence about why it should stand. A person — not the automated screen — reviews it, and we reply within five business days. If we agree, the model is restored and any credits charged are returned.

Changes

We update this page when our rules or our screening change. The date at the top is the last revision.