Privacy Policy

Last updated: 2 August 2026

FactCheck ("the service", "we", "us") is an automated messaging assistant you can message directly on Instagram (@check.this.for.me), WhatsApp, or Telegram. You send it something you saw online — a reel, a post, a forwarded message, a link, or a question — and it researches and replies with a verdict and its sources. This policy explains exactly what we collect, why, and how to make us delete it.

What we collect

Only what you send us directly, in your conversation with the bot:

• Your platform-assigned identifier (Instagram-scoped ID, WhatsApp phone number, or Telegram chat ID) — this is how we know which conversation is yours. We do not receive your name, profile photo, contacts, or anything else from your account.
• The text of messages you send the bot.
• Any media you share with the bot (photos, videos, reels/posts) — downloaded temporarily to our server so it can be analysed.
• Facts you explicitly tell the bot about yourself in conversation (e.g. "I'm applying for a UK visa"), which it may remember to give you better answers later. We do not infer, scrape, or purchase anything about you from any other source.
• Basic technical metadata: message timestamps, which channel you used, and coarse usage stats (e.g. how many checks you've run) for abuse prevention and to improve the service.

We do not look at your public profile, followers, posts, or any content beyond what you send us in the chat. We do not read DMs you have with anyone else — Instagram and WhatsApp only ever deliver us messages sent directly to this bot.

Why we collect it

• To research and fact-check the content you send, and cite our sources back to you.
• To hold a coherent conversation across multiple messages (so you can ask follow-up questions about something you sent earlier).
• To personalise responses using facts you've told us (optional — you never have to share anything personal for the core fact-checking to work).
• To prevent abuse (rate limits) and to understand, in aggregate, what kinds of content people are checking, so we can improve the service.

Who else sees your data

To generate a response, your message and any shared media are sent to one or more of the following processors, in this order of preference, purely for the purpose of producing an answer:

• NVIDIA (NIM API) and OpenRouter (an API marketplace that routes to one of several underlying AI model providers) — used to understand your message and draft replies.
• Perplexity AI — used only to help find candidate web sources; we independently fetch and verify pages before citing them, we never take Perplexity's summaries as fact.
• A self-hosted search index (SearXNG) running on our own server — this does not send your query to any third party.
• Audio transcription (for videos you share) runs locally on our own server, not on any external service.

These providers process data under their own standard API terms, which as of this writing state that data submitted via API is not used to train their models — but we do not control their infrastructure and cannot make guarantees on their behalf beyond what each provider publishes. We do not send your data to advertisers, data brokers, or anyone for marketing purposes, and we never sell your data.

Where it's stored, and for how long

Conversation history, remembered facts, and temporarily cached media are stored as plain files on our own server. Cached media (the actual video/image files you shared) is automatically deleted after 7 days regardless of anything else. Conversation history and remembered facts are kept until you ask us to delete them (see below) — we do not currently auto-expire this data, so if you want it gone, tell us.

Your rights, and how to delete your data

You can ask the bot to forget you, on any channel, at any time. See the Data Deletion Instructions page for the exact steps — it takes one message and is immediate and permanent.

You can also ask what we have stored about you, or object to any use described here, by emailing .

Children

This service is not directed at children and we do not knowingly collect data from anyone under 13 (or the minimum age required by your platform's own terms, if higher). If you believe a child has used this service, email us and we will delete the data.

Security

Data is stored on a server we control, is not publicly accessible, and is never committed to any public code repository. As with any online service, we cannot guarantee absolute security, but we do not take backups off-server or replicate this data anywhere else.

Changes to this policy

If this policy changes materially, we will update the date at the top of this page. Continuing to use the service after a change means you accept the update.

Contact

This service is operated by an individual developer/small business (GST-registered, India; full details on the Business Information page). For any question, request, or concern: .