Fake Social Profile Generator

Fake social media profiles for UX, demos and design systems

A fake social profile bundles avatar, username, bio, location, follower count and a sample post into a coherent persona. The point is to populate UI mockups, Storybook fixtures and ads previews without scraping real accounts. Real users — even public ones — fall under platform terms of service and, in Brazil, under the LGPD; a synthetic profile sidesteps both, as long as you do not use it to impersonate a living person.

Typical legitimate use cases: wireframing a Twitter/X clone or Instagram timeline, populating Buffer/Hootsuite-style dashboards for sales demos, training social-media managers on a sandbox, generating creative previews for Meta Ads Manager, and feeding QA scripts for a community feature without violating Meta's automation policies.

Anatomy of a synthetic profile

• Avatar — three reliable sources: ui-avatars.com (initial-based fallback), procedurally generated SVG (DiceBear's avataaars, bottts, thumbs), and AI portraits from this-person-does-not-exist.com (StyleGAN2 — fully synthetic faces).
• Username — Reddit-style adjective + noun + number (happy_panda42) is the most natural-looking pattern.
• Bio — emoji-heavy, with role identifier, hashtag and "link in bio" — that is the format the algorithm rewards on TikTok and Instagram.
• Follower count — long-tail distribution (most accounts under 1k, a few influencers over 100k); never spread evenly.

Where it crosses the line

Creating fake accounts at scale on a real platform violates Meta, X (Twitter) and TikTok terms of service — Meta's Quality Score, X's automation detection and Cloudflare Turnstile / hCaptcha challenges exist precisely to catch this. The TikTok BR Transparency Report 2024 reports 12 million fake accounts removed in Q1 alone. Botometer (Indiana University) is a public benchmark for bot detection on X.

Worse, impersonating a real person — even with a StyleGAN face — can be charged as falsa identidade (art. 307 of the Brazilian Penal Code) or as estelionato if any economic damage is caused. For development you should restrict synthetic profiles to your local mock server or to platform sandboxes (Meta's Graph API has a sandbox mode for exactly this).

Data-protection note

Synthetic data is not personal data under the GDPR or the Brazilian LGPD (Lei nº 13.709/2018) — you can share, publish or commit the output freely. The moment you mix real handles, real avatars or real posts into the dataset, the regime changes and you need a lawful basis. Keep the boundary clean.

FAQ

Can I use these profiles in a UX case study? Yes. Synthetic personas are the standard way to illustrate flows in a portfolio without revealing real customers.

Can I open a real account on Instagram with the generated data? No. That violates Meta's terms of service and, if a real person is impersonated, it may amount to identity fraud.

Are the AI-generated faces of real people? StyleGAN portraits are fully synthetic — no real subject — but you should still avoid using them to claim someone exists for advertising purposes.

Can I generate one avatar at a time? Yes. Use ui-avatars.com/api/?name=John+Doe or DiceBear's URL API for a single deterministic SVG.