Acceptable Use Policy

1. Scope

This policy applies to every piece of content that passes through the AI customization feature:

• Text — descriptions and prompts you submit;
• Images — photos you upload, and the keyframe images generated from them;
• Video — the animation clips generated for your custom pet.

The standards apply to inputs and outputs alike: a description, an uploaded photo and the generated pet must all comply. DeskSprite is a desktop pet app for general audiences — if content would be unsuitable there, it is unsuitable here.

2. Content standards: prohibited categories

You must not upload, describe, prompt or generate content in any of the following categories. Violations lead to the enforcement actions in section 7.

1. Sexual / NSFW content — pornography, explicit nudity, sexual acts or fetishes, and any content whose primary purpose is sexual arousal.
2. Violence and gore — graphic violence, gore, mutilation, torture, or cruelty to animals. DeskSprite pets are cute companions; violent content has no place in the product.
3. Hate speech and harassment — content that attacks, demeans, dehumanizes or incites hatred or violence against people based on protected characteristics (such as race, ethnicity, national origin, religion, disability, gender, gender identity, age or sexual orientation), or that harasses or bullies identifiable individuals.
4. Child safety (CSAM) — any sexualization of minors, child sexual abuse material, or content that endangers children in any way. This is an absolute ban: suspected violations are reported to the relevant authorities and the account is permanently banned (see section 2.1).
5. Deepfakes and impersonation — photorealistic depictions of real, identifiable people (including public figures) without their documented consent, content that misrepresents what a real person said or did, or impersonating another person, brand or organization.
6. Copyright and trademark infringement — reproducing protected characters, artworks, logos, brands or other intellectual property that you do not own or have no license to use.

Content that is illegal under applicable laws — including the laws of China, where the service is operated — is likewise prohibited.

2.1 Zero-tolerance content

2.2 No restricted-permitted tiers

DeskSprite does not operate an adult/18+ tier and does not use age verification, so there is no context in which the six categories above become permitted. The service also does not produce medical, legal or financial guidance content.

2.3 Other content rules

• No disinformation designed to deceive about elections, public health or other matters of public interest;
• No presenting AI-generated content as genuine human work to deceive;
• No bulk spam generation or automated circumvention of quota rules;
• No attempts to bypass or "jailbreak" the content filters described in section 3.

3. Content moderation

We moderate with layered controls, combining automated filtering and human review:

• Automated filtering at generation time (real-time) — generation runs on third-party AI models (image generation via the gpt-image-2 model accessed through API service providers; video generation via Seedance models on ByteDance's Volcano Engine). These models run their own built-in content safety filters that detect and refuse prohibited inputs and outputs; refused jobs fail without delivering content, and the customization quota is returned to your balance.
• Input validation on our side — uploads pass server-side checks (file type, size and count), and submissions that our providers' filters reject are blocked from delivery.
• Human review — the operator personally reviews reported or flagged custom pets and samples delivered content on a regular basis (at least monthly). Content found to violate this policy is removed and not delivered.

Violations are triaged by severity:

4. Reporting mechanism

• What you can report — any custom pet or submitted content that you believe violates this policy, regardless of whose account generated it.
• What to include — the pet's name (as shown in the shop or app), a screenshot or link if you have one, what rule you believe was broken, and your contact email if you want follow-up.
• In good faith — report what you genuinely believe violates this policy. Deliberately false mass-reporting to harm another user is itself a violation.

5. How to report

6. Response times

Process for every report: we review the reported content → decide whether it violates this policy → take the enforcement action (section 7) → notify you of the outcome where you provided contact details.

7. Enforcement actions

Depending on severity, we may take one or more of the following:

• Refuse generation or delivery — a violating submission is blocked; refused jobs do not consume quota;
• Remove content — a delivered custom pet that violates this policy is removed from your account and the shop;
• Restrict features — for example, disabling the AI customization feature for the account;
• Suspend the account — temporarily, pending review of serious or repeated violations;
• Permanently ban the account — for zero-tolerance content (section 2.1) or repeated serious violations;
• Refer to authorities — where the law requires or allows it.

8. Appeals

• You may appeal any enforcement action within 14 days, via [email protected] (subject "Appeal") or the Feedback page, including your account email, the order number if applicable, and why you believe the action was wrong.
• Actions triggered by automated filtering are always re-reviewed by a human (the operator) on appeal.
• We decide appeals within 10 business days. If the appeal succeeds, removed content is restored where possible and feature restrictions are lifted.

9. Changes to this policy

We may update this policy as the service and applicable rules evolve. Material changes will be announced on this page at least 15 days before they take effect, with a new version number and effective date (consistent with the Privacy Policy).

10. Contact