Skip to main content

AI Content Policy

What you may and may not generate with our AI models, how we moderate content, and how to report abuse

Version 1.0 · Effective 27 October 2026 · Part of our Terms of Service

1. Scope

This policy explains what our AI features may and may not be used to create, how we moderate generated content, and how anyone can report content or behaviour that breaks these rules. It applies to every AI feature in the Service — text, image, audio, and video generation, chat, image captioning, and any output produced through our API.

It sits alongside, and is part of, our Terms of Service (in particular section 10, Acceptable use). Where this policy and the Terms disagree, the Terms win.

2. Prohibited content

You must not use the Service to create, upload, request, or distribute content in the following categories. These rules apply to your prompts, to files you upload for processing, and to anything you generate.

2.1 Sexual and NSFW content

  • Pornographic, explicit, or sexually suggestive material.
  • Sexual content involving minors, or anything that sexualises a person who appears to be under 18.
  • Non-consensual sexual content, including sexual imagery of a real person created or altered without their consent ("deepfake pornography").
  • Fetish content whose primary purpose is sexual gratification.

2.2 Violence and gore

  • Graphic depictions of violence, injury, dismemberment, torture, or death.
  • Content that glorifies or instructs on committing violent acts.
  • Animal cruelty, mutilation, or realistic gore intended to shock.
  • Content that promotes self-harm, suicide, or eating disorders.

2.3 Hate speech and harassment

  • Content that attacks, dehumanises, or incites hatred against people based on race, ethnicity, nationality, religion, sex, gender identity, sexual orientation, age, or disability.
  • Slurs used to demean a protected group.
  • Targeted harassment, bullying, threats, doxxing, or coordinated abuse of a specific person.
  • Content that promotes terrorism or violent extremism.

2.4 Content that is unsafe for minors

  • Child sexual abuse material (CSAM) in any form, including drawings, illustrations, or text describing it. We report this to the relevant authorities and close the account permanently.
  • Content that grooms, exploits, or endangers minors.
  • Content designed to expose a minor to sexual, violent, or otherwise harmful material.

2.5 Deepfakes, impersonation, and misleading synthetic media

  • Audio, images, or video that impersonate a real, identifiable person, whether public figure or private individual, in a way that could deceive.
  • Content presenting a real person saying or doing something they did not say or do.
  • Material that falsely implies endorsement by a person, brand, or institution.
  • Content intended to influence an election, a financial decision, or public safety with fabricated evidence.
  • Reproducing or closely imitating a copyrighted work, character, artwork, song, voice, or performance without the rights holder's permission.
  • Using another party's trademark, logo, or brand identity in a way that suggests affiliation, endorsement, or authorised resale.
  • Reproducing a living artist's signature style for commercial distribution where that reproduces their protected works.

2.7 Other illegal or harmful uses

  • Malware, phishing material, fraud, scams, or social-engineering scripts.
  • Content that violates privacy law, discloses personal data, or exposes confidential information.
  • Medical, legal, or financial advice presented as the judgement of a licensed professional.
  • Anything that breaks applicable law, regulation, or card network rule.

3. How these rules apply in practice

  • Prompts count. Asking for prohibited content is itself a breach of this policy, even if the model refuses or the output is unusable.
  • Uploaded inputs count. Sending a prohibited image or document to the Service for processing is a breach, just like generating one.
  • Attempts to evade count. Rewriting a prohibited prompt in euphemisms, another language, encodings, or multi-step workarounds is a breach.
  • No obligation to pre-screen. We are not required to review your content before you use it, and we do not guarantee that every prohibited output will be blocked. You remain responsible for what you generate and publish.

4. Moderation

We combine upstream model safety filters with our own safeguards:

  • Model-level filtering. Requests are routed to Cloudflare Workers AI, OpenRouter, OpenAI, or Anthropic according to your model choice. Each provider applies its own safety systems, which can refuse or block a request before it reaches you.
  • Service-level safeguards. We apply input checks, rate limits, abuse detection, and account-level blocking on top of provider filtering.
  • Human review. Reports, appeals, and suspected abuse of the rules above are reviewed by our team. Automated decisions that remove access can be appealed (section 7).

5. Reporting content or behaviour

If you find content generated through the Service, or a user of the Service, that breaks this policy, tell us:

  • Email: support@vibestrap.dev
  • Subject line: AI content report
  • Include: what you found, where it is (link or account identifier), when you saw it, and any screenshot or reference that helps us locate it.
  • Copyright or trademark claims: use the same address with the subject IP takedown, and include the work or mark you own, the material you say infringes it, and a statement that you are authorised to act for the rights holder.
  • Do not send us the illegal material itself. Describe it. We do not want copies of prohibited content, especially anything involving minors, in our inbox or systems.

We acknowledge every report within 3 business days. Reports involving apparent child sexual abuse material, credible threats to safety, or non-consensual intimate imagery are treated as urgent and acknowledged within 24 hours.

6. What happens after a report

  1. Acknowledge and triage. We log the report and assign a severity based on the category in section 2.
  2. Contain. Where there is an immediate risk, we restrict the reported content or the associated account while we review.
  3. Review. We assess the report against this policy, the Terms, and applicable law, and may ask you for further detail.
  4. Act. Depending on the finding, we may take no action, issue a warning, remove the content, forfeit unused credits tied to the prohibited generation, suspend the account, or terminate it. Serious cases under sections 2.4 and 2.5 result in immediate termination.
  5. Notify. We tell the reporter and the account holder what we decided, unless doing so would compromise an investigation, endanger someone, or breach the law.
  6. Refer. Where the law requires it, we report to law enforcement or the relevant child-protection authority and preserve the associated records.

We keep a record of reports, decisions, and actions for as long as needed to enforce this policy and to meet legal obligations.

7. Appeals

If your account or content was restricted under this policy and you believe we got it wrong, reply to the decision email or write to support@vibestrap.dev within 30 days with the context you think we missed. A second reviewer re-examines the decision. Termination for child sexual abuse material is not appealable.

8. Changes

We may update this policy as our features, our providers, or the law change. Material changes are announced in the Service or by email at least 15 days before they take effect, and the version and effective date at the top of this page always reflect the current text.

9. Contact

Questions about this policy, or about a moderation decision: support@vibestrap.dev. Postal contact details are listed in section 16 of our Terms of Service.