Last updated: November 4, 2026
SocialFansGeek AI Studio helps creators make photos, posters, short videos, clips and captions faster. This page explains how our AI works, how we label it, and what we will not make.
1. Our principles
- People first: AI helps creators; it does not pretend to be them or anyone else.
- Honest by default: AI-made content is labelled, and we never present AI output as real footage.
- Consent for likeness: you can only transform photos, video or voice of yourself or people who agreed.
- Privacy: uploads are used only to make your result, deleted on a schedule, and never used to train models.
- Human oversight: we review reports, test new models before switching them on, and can switch off any tool that is misused.
2. Which tools use AI and which models
| Tool | What AI does | Model type |
|---|
| AI photo looks and trend templates | edits or restyles your photo | image generation models via OpenRouter |
| Posters and text-to-art | creates images from your text | image generation models |
| Photo to video | animates an image into a short clip | video generation models (such as Kling and Google Veo) |
| AI Clips | picks the best moments from your video and cuts clips | Google Gemini picks moments; cutting runs in your browser |
| Video Analyzer | scores a video and suggests improvements | Google Gemini |
| Free AI text tools | captions, hashtags, bios, scripts | large language models |
We list model families, not a promise of a fixed model; we may change models to improve quality or safety.
3. How we label AI content
- We are adding a machine-readable marker to results (file metadata, and C2PA Content Credentials where the model provides them), as the EU AI Act asks of generative AI.
- We are adding an "AI-generated" note to AI Studio results and an option for a small visible label on video exports.
- When you post AI content that looks like a real person or event, label it. TikTok, Instagram, YouTube and Facebook all have an "AI-generated" label for this.
- Text from our AI writing tools is a draft for you to edit. If you publish AI text on a matter of public interest without human review, say it was AI-assisted.
4. What we will never generate
- Sexual or nude images of real people, or any sexual content involving minors.
- Deepfakes of real people without their consent, including celebrities and politicians.
- Content meant to deceive voters, fake news events, or impersonate public authorities.
- Fake IDs, documents, receipts or reviews.
- Content that harasses, threatens, or promotes hate or violence.
- Content that copies someone else's protected characters, logos, artwork or music.
- Anything that uses AI in ways the EU AI Act prohibits, such as manipulative techniques that exploit vulnerable people, social scoring, or inferring emotions or sensitive traits from faces.
Our model providers also run safety filters. We block or remove content and close accounts that break these rules.
5. Our AI and search engines
We publish AI-assisted content on our own site only when a person has reviewed it and it is useful to readers. We do not mass-produce pages to rank in search, and we follow Google Search's spam policies on scaled content.
6. Report misuse
If you see content made with our tools that shows you without consent, is sexual, or breaks this policy, email support@socialfansgeek.com with the subject "Abuse report" and the link. We remove intimate images and deepfakes within 48 hours of a valid report, and we respond to other reports within 3 business days.
7. Questions
Ask about AI at support@socialfansgeek.com. We review this policy at least twice a year as laws and our tools change. See also our Terms of Service and Privacy Policy.
.png)
.png)
