한국어Get started

Draft

Privacy Policy

Drafted 2026-10-21 · The effective date is set at launch

This is a draft that has not been reviewed by a lawyer yet. It may change before launch.

Nexframe (넥스프레임, “we”) runs the hydek app and the hydek.io website (the “Service”). This policy explains what personal information we process, why, and the choices you have. It follows the Korean Personal Information Protection Act (PIPA).

1. In short

  • hydek is an app that runs on your computer. Your video files and results stay on your computer by default.
  • Sign-in for the AI you use (ChatGPT or Claude) is handled on your computer by that service’s official program. We never receive or store those credentials.
  • Payment details such as card numbers are handled by our reseller, Paddle. We never receive them.
  • This website has no ad trackers and no analytics scripts.

2. What we process, why, and for how long

Category Data Purpose Retention
Sign-up and sign-in (Google account) Email address, name, profile picture, Google account ID Create your account, verify you, support Until you delete your account
Subscription and payment Paddle customer ID, transaction ID, subscription status, country, amount and currency Provide your plan and credits, refunds, tax 5 years, as required by Korean e-commerce law
Usage records Credit usage, records of tasks that used our servers (time, status, cost), app version, operating system Credit accounting, troubleshooting, improving the Service Until you delete your account; billing records 5 years
YouTube connection (if you connect it) YouTube channel ID and name, IDs and URLs of videos you upload, OAuth access token Upload the videos you approve to your channel and confirm the result Until you disconnect or delete your account
Server runs, cloud speech recognition (if you choose them) The files and text a task needs (audio from your video, captions, instructions, generated text) Run the task you asked for 30 days after the task finishes
hydek bot (if you choose it) Instructions, file content a task needs, AI output Run the task you asked for Content is passed through, not stored. Request time, token count and cost are kept as usage records
Problem reports (if you consent) App logs, error details Find the cause of a problem 90 days
Support requests Email address, message Answer you 3 years
Access logs IP address, time of access Security and abuse prevention 3 months

We do not use personal information for any other purpose without your consent.

3. Google user data

hydek uses Google Sign-In and the YouTube Data API.

  • From Google Sign-In we receive your email address, name and profile picture, and use them only to create your account and verify you.
  • We use YouTube access only to upload videos you approve to your own channel and to confirm the upload. Right before an upload, hydek asks you to approve it and choose its visibility. If you approve a channel and visibility ahead of time when you create a workflow, hydek uploads only within that choice without asking again.
  • We do not use information from Google APIs for advertising, we do not sell it, and no person reads it, except when you ask us to, or when security or the law requires it.
  • hydek’s use and transfer to any other app of information received from Google APIs will adhere to the Google API Services User Data Policy, including the Limited Use requirements.
  • hydek uses YouTube API Services. By using hydek’s YouTube features you agree to the YouTube Terms of Service, and Google’s handling of your data is described in the Google Privacy Policy.
  • You can revoke hydek’s access at any time from the app’s connections screen or from your Google Account security settings. When you do, we delete the tokens we hold without delay.

4. Processors and transfers abroad

We entrust the processing of personal information to the companies below to provide the Service. They are outside Korea, so the data is transferred abroad (PIPA Article 28-8(1)(3)). Transfers happen over encrypted connections (TLS) when you use the related feature.

Recipient (contact) Country Data Purpose Retention
Cloudflare, Inc. (privacyquestions@cloudflare.com, privacy policy) United States and countries where Cloudflare runs data centers Account, payment and usage records in section 2, request data and IP address, task content in transit Servers, storage, website hosting As in section 2
Paddle.com Market Limited (privacy@paddle.com, privacy policy) United Kingdom Email address, country, payment and billing details Reseller (Merchant of Record), payments, tax, refunds Per Paddle’s policy
Google LLC (googlekrsupport@google.com, privacy policy) United States Sign-in data, videos you approve with their title and description Google Sign-In, YouTube uploads Per Google’s policy
OpenRouter, Inc. (169 Madison Avenue, New York, privacy policy) United States Instructions, file content a task needs, AI output, hashed user ID Model routing for hydek bot Deleted after the output is generated; request metadata is kept
Deep Infra Inc. (policy@deepinfra.com) United States Same as above AI processing for hydek bot Deleted right after processing
Baseten Labs, Inc. (privacy@baseten.co, privacy policy) United States Same as above AI processing for hydek bot Per the provider’s terms (being confirmed)
Amazon Web Services, Inc. (aws-korea-privacy@amazon.com, privacy policy) United States Same as above AI processing for hydek bot Processed with retention turned off
  • Onward processing. OpenRouter passes hydek bot requests to the AI processing providers in the table above (Deep Infra, Baseten). OpenRouter may classify requests automatically and anonymously to prevent abuse.
  • No training. We configure hydek bot requests to go only to providers that refuse data collection and keep no data, and the providers do not use these requests to train models. OpenRouter deletes request content after generating the output and keeps only metadata such as time and usage.
  • We will add the cloud speech recognition provider to this table before that feature opens.
  • How to refuse, and what happens. If you do not use hydek bot or cloud speech recognition, or you connect your own ChatGPT or Claude, nothing is transferred to the AI processing providers, and hydek bot is not available. Payments and servers (Paddle, Cloudflare) are essential to the paid Service, so refusing those transfers means you cannot use the paid Service.

We do not share personal information with third parties except as above or where the law requires it.

5. Deletion

When a retention period ends or the purpose is met, we delete the data without delay. Electronic files are deleted so they cannot be recovered. Data we must keep by law is stored separately and only for the required period.

6. Your rights

You can ask to access, correct, delete or stop the processing of your personal information, or withdraw consent, at any time. Do it in the app’s account screen or email privacy@hydek.io, and we will reply within 10 days. A legal representative or someone you authorize can also make the request. You must be 19 or older to sign up.

7. Security

  • All traffic is encrypted (HTTPS).
  • Sign-in secrets are stored as hashes, and access to personal information is limited to the people who need it.
  • We do not store payment details.
  • We keep access logs and review unusual access.

8. Cookies

This website does not use cookies for tracking or advertising. Our host (Cloudflare) may set strictly necessary cookies for security. The checkout is provided by Paddle and follows Paddle’s cookie policy.

9. Privacy officer

  • Officer: Younsu Kim
  • Email: privacy@hydek.io

You can also ask these Korean public bodies for counsel and remedies:

10. Changes

We post changes on this website at least 7 days before they take effect. For changes that are unfavorable to you, we give 30 days’ notice and also email you.