Acceptable Use Policy — AI-Generated Content

1. Scope

• Uploads: the photo you upload to animate.
• Video: AI-generated short videos (MiniMax H3 via fal.ai), with silence or model-synthesized sound.
• Inputs: template choice and file names. Users cannot type free-text prompts; every template prompt is fixed on our server and reviewed by us.

2. Content standards & prohibited categories

You may not upload, generate, or share any of the following six categories of content:

1. Pornography / NSFW — nudity, sexual or sexually suggestive content of anyone.
2. Violence / gore — injury, blood, killing, torture, or glorifying violence.
3. Hate speech — content attacking people for protected characteristics, hate symbols, extremist imagery.
4. Child-unsafe content (CSAM) — any image of a minor (under 18), and any sexual content involving minors.
5. Deepfakes / impersonation of real people — any person other than yourself (or someone who gave you explicit permission), and in all cases celebrities, public figures and politicians; content meant to deceive anyone into thinking a real event happened.
6. Copyright / trademark infringement — copyrighted characters, brands, logos, or other people's artwork, and claiming you own copyrighted music added to an output.

2.1 Absolutely prohibited — zero tolerance

Actioned immediately and, where required, reported to the authorities:

• Child sexual abuse material or any sexual content involving minors.
• Content that promotes or facilitates terrorism, extremism, or mass violence.
• Instructions for weapons of mass destruction.
• Content inciting genocide or ethnic hatred.
• Deepfakes of real individuals — face swaps, nudification, intimate-scene compositing (e.g. kissing/hugging two real people), and fabricated statements or actions intended to pass as real — whether or not the people involved consent. Ideoy offers none of these features.

2.2 Not offered on Ideoy

Ideoy does not allow adult content, realistic violence, political content or advice of any kind, so there is no "restricted with review" tier and no age-gated content. All users and everyone shown in uploads must be 18+.

2.3 Generally prohibited conduct

• Using outputs to spread misinformation, or presenting an output as real footage.
• Impersonating real individuals, brands, or organizations.
• Infringing intellectual property, publicity, or privacy rights.
• Removing or hiding the AI-generated nature of outputs, or passing them off as human-made or real.
• Spam, bulk account creation, or automated abuse of paid generation quotas.
• Trying to bypass our safety checks (e.g. disguised files, renamed files, cropping to evade review).
• Templates are unofficial, fan-made takes on internet meme trends and are not affiliated with or endorsed by the original creators, performers, or rights holders.

2.4 Source of uploaded material

You may only upload your own photos, or material for which you have explicit permission from the people depicted. Uploading another person's photos without their consent is prohibited. By starting a remake you agree the photo is of you (or you have permission) and that everyone shown is 18+ — we store a timestamp on every generation. Permission does not authorize generating the deepfake content prohibited under Section 2.1.

Every Ideoy output is a new AI-generated video of the person in the uploaded photo performing a fictional meme scene. Outputs are always labelled AI-generated on the site; when you share one, keep it clearly identified as AI-generated and enable the posting platform's AI-generated content label (e.g. TikTok, Instagram, YouTube). Never present an output as real footage.

3. Content moderation

We use automated checks before and after generation, plus human review of everything flagged or reported.

3.1 Automated review

• Fixed prompts: users cannot write prompts. Template prompts are written and reviewed by us and checked against our blocklist before release.
• Before generation: Terms / Acceptable Use agreement on remake; real file-type check (images only, ≤5MB); a versioned blocklist (current version 2026-10-20.1) covering NSFW, child-safety, violence, hate, self-harm, deepfake/celebrity/political-figure and copyrighted-character/brand terms; and a vision model check of the photo for possible minors and public figures. Matches are refused before any video is generated.
• Model-side: MiniMax H3 via fal.ai applies fal / MiniMax content-safety checkers.
• After generation, before display: a frame of every generated video is checked by a vision model for NSFW, violence/gore, possible minors, public figures and hate symbols. Flagged results are not saved or shown, and any credit used is returned. If the automated check cannot run, that is logged and the result is included in human review.
• Usage limits: per-user and per-IP daily caps, bot challenge (Cloudflare Turnstile) and rate limits.
• Safety models used: LLaVA 1.5 7B on Cloudflare Workers AI, our own blocklist, and fal / MiniMax safety checkers.

3.2 Human review

• Every report, and every automatically blocked output, goes to a review queue checked by the founder.
• Ideoy is run by one person; there is no 24/7 team. Response times are in Section 6.

3.3 Periodic audits

• We review the blocklist and moderation logs at least every two weeks and whenever a template is added; changes are versioned.
• Every new template is checked before launch: no real people, celebrities or copyrighted characters.

3.4 Severity levels

4. Reporting

• Report anything that breaks this Policy, especially anything involving a minor, and any output that uses your likeness without your permission.
• Confidentiality: we never reveal who made a report.
• Impartiality: each report is reviewed on its merits.
• Anti-abuse: malicious or false reports are recorded and may be actioned.

5. Reporting channels

• Online report form: ideoy.ai/report (stored in our database and emailed to the founder's monitored inbox).
• In-product: a Report button on every finished video, in My videos, and on every sample in the Feed.
• The same form handles support questions and appeals.

6. Response timelines

• L1 (child safety / terrorism): the content is taken down within 24 hours of the report.
• Other reports: reviewed within 3 business days.
• You receive an email reply when there is a decision.

7. Enforcement

These are the actions we can and do take:

• Refusal: blocked uploads and outputs are never generated or shown.
• Content removal: we delete the video from storage and revoke its download.
• Warning by email.
• Account suspension or permanent ban: a suspended account cannot generate or buy.
• Credit forfeiture: for L1/L2 violations, remaining credits are forfeited without refund.
• Legal reporting: CSAM and other criminal content is reported to the relevant authorities.

Every enforcement decision is logged. Payment refunds and chargebacks also remove the credits bought with that payment (see Refunds).

8. Appeals

• If you think a block or enforcement action was wrong, appeal within 30 days at ideoy.ai/report and choose "Appeal a moderation or enforcement decision". Include the generation id if you have it.
• We reply within 10 business days. If the decision was wrong we restore access and return credits.
• Automated false positives happen (for example, an adult flagged as a possible minor). Retrying with a different clear photo of yourself is fine.

9. Moderation records & policy updates

• We keep moderation records (automated results and human decisions) for 12 months, and reports and enforcement decisions for 24 months, so we can handle appeals and answer our payment provider or authorities. Uploaded photos themselves are deleted after about 30 days (see Privacy).
• Material changes to this Policy are announced on this page at least 15 days in advance. Continued use means acceptance.

10. Contact

• Safety reports, appeals and support: ideoy.ai/report
• Operator: independent individual developer, Ideoy · ideoy.ai