Croche

Croche Chat Privacy Policy

Beta v0.1 · Effective October 14, 2026 · Last updated October 14, 2026

This is the beta version that applies during the Croche beta. Before general availability it will be replaced by a version reviewed by legal counsel, with advance notice under the change procedure in this document.

HIRE AI Inc. (operating as H1R AI; "we") processes personal information of Croche Chat ("the app") users as follows. This policy covers the app only.

Summary. No sign-up or login. We don't collect names, emails, phone numbers, location, contacts or photos. Conversations and Moongchi's memory are stored on your device; our servers don't keep them. To generate replies, your messages and conversation context go from the app through an AI routing service (OpenRouter) directly to AI model providers; our servers don't receive conversation content. No advertising, analytics or crash-reporting SDKs.

What we process.

  • On your device only: conversations, Moongchi's memory (conversation summaries), room state, coin balance, and language. These are kept until you delete the app or its data.
  • Sent to the AI routing service, not to our server: your messages and the conversation context needed for a reply, used to generate the reply. We don't keep them (see processors below).
  • Stored on our server: a random identifier created at install, plus usage records (time, duration, model, result, cost). These are used for operations, abuse prevention, cost accounting and troubleshooting. They are kept for one year.
  • Device integrity: Google Play Integrity verdicts and failure reasons, used to prevent abuse.

Processors and international transfers. We use the following processors, which involves transferring data abroad:

  • Cloudflare (US and other data-center locations): hosting.
  • OpenRouter (US): routing requests to AI models; prompt logging is off.
  • AI inference providers reached through OpenRouter (the US, China and other countries where each provider is based): generating replies. We exclude providers that use inputs for training. Providers may retain data briefly for caching or abuse prevention under their own policies.
  • Google (US): Play Integrity.

AI replies. Replies are AI-generated and may be inaccurate or inappropriate. Tell us about inappropriate replies at the contact below; we use what you send only for review and improvement and delete it after 90 days. We don't use your conversations to train AI models, and we don't send them to providers that train on inputs.

Your rights. You may ask to access, correct, delete, or stop the processing of your data. Because server records aren't linked to your identity, please tell us roughly when you used the app and on what device, and we will act on what we can find. Conversations live only on your device, so clearing the app's data deletes them at once.

Children. The app isn't directed to children under 14.

Contact. Privacy officer: Simyung Jang (CEO), privacy@h1r.ai. HIRE AI Inc. (Business Reg. No. 141-86-03506), 2F 207 (No. 23), 10 Gwanggyojungang-ro 296beon-gil, Suji-gu, Yongin-si, Gyeonggi-do 16943, Republic of Korea.