Privacy Policy
1. Plain-English summary
Signal is an AI-powered tool that reads social media comments on accounts you own, classifies them by importance, and alerts you to the ones that need attention. To do that, we need to:
- store your account info (email, name) so you can sign in;
- receive comments and mentions from the social platforms you connect (YouTube, Instagram, Facebook);
- send each comment to an AI service (Anthropic Claude) to classify it;
- save the results so you can read them in the dashboard.
Signal never posts, replies, or acts on your behalf. It reads comments and recommends; a person on your team decides what to do. We do not sell your data, run advertising on it, or train AI models on it. The full details are below.
2. Who we are
The data controller for Signal is Nevcan Uludaş, a natural person. Turkish data protection law (KVKK Art. 3) and the GDPR both define a controller as a natural or legal person, so there is no company registration number to give. For questions about this policy, or to exercise any of the rights described below, contact us at signal@getpeyk.com.
Signal is published as Peyk Signal, one of two independent products under the Peyk brand. Peyk is a brand, not the controller: the other product has its own separate controller and its own privacy policy, and this policy covers Signal only. If you are looking for the other product's policy, this is not it.
3. What we collect
3.1 Account data
- Email address, first and last name, phone number (optional)
- Telegram chat ID (optional, only if you choose to receive alerts there)
- Authentication metadata managed by Firebase Authentication (sign-in time, password reset events)
3.2 Workspace and configuration data
- Workspace name, organization names, channel labels you create
- API keys you provide for the platforms you connect (YouTube, Meta) — see Security for storage details
- Threshold and rule settings you configure
3.3 Social media content (the comments themselves)
When you connect a YouTube channel, an Instagram account or a Facebook Page you own, Signal receives the comments and mentions posted on that account by other people. For each, we store:
- The text of the comment
- The author's username on the source platform
- A link back to the original post
- What Signal derives from it: the AI classification (a one-line summary, the reason for the classification), our priority score, and a suggested reply for your team
The author of a comment is a user of the platform the comment was posted on, not a user of Signal. We process their public comments only because you, the operator of the receiving account, have asked us to. If a comment author wants their data removed, see Your rights.
3.4 Cookies
| Cookie | Purpose | Lifetime |
|---|---|---|
__session | Firebase Authentication session — proves you are signed in | 5 days |
sm_workspace | Remembers which workspace you last selected | 30 days |
We do not use third-party advertising or tracking cookies.
4. Third parties
4.1 Where the data comes from
Comments reach Signal from the platforms you connect. These platforms are sources, not parties we share data with: we collect from them, and we send them nothing but the credentials needed to authenticate. Signal never posts, replies, or acts on your behalf on any platform.
| Source | What we receive |
|---|---|
| Meta (Instagram, Facebook) | Comments and mentions on the Instagram accounts and Facebook Pages you connect, delivered by webhook |
| YouTube Data API (Google) | Comments on the YouTube channels you connect |
4.2 Who we share it with
We share data only with the parties below. Each receives only what its function needs:
| Provider | Purpose | What we send |
|---|---|---|
| Google Cloud (Cloud Run, Firestore) | Hosting and database | Everything stored in the service |
| Firebase Authentication (Google) | Sign-in | Your email, name and sign-in data |
| Anthropic (Claude API) | AI classification of each comment | The comment text, the platform it came from, and our instructions to the model. The author's username is not sent. Anthropic does not train on API inputs by default. |
| Telegram (Bot API) | Alerting the members of your workspace who opt in | The recipient's Telegram chat ID and an alert carrying only a priority level, a score, the platform, the date and time, and the comment's intent. No comment text and nothing about its author. |
We do not sell your data, share it with advertising partners, or use it to train any AI model.
5. How we use the data
- To run the service — sign you in, receive your social comments, classify them, save them, alert you.
- To improve reliability — operational logs (HTTP requests, errors) are kept by Google Cloud Run for diagnostic purposes. These do not contain comment text by default.
- To respond to you — when you contact support.
We do not use the data for marketing, advertising, or to profile end users.
5.1 Legal basis
| Whose data | What | Legal basis |
|---|---|---|
| Signal users (you and your team) | Account data, Telegram chat ID, workspace configuration | Performance of a contract: GDPR Art. 6(1)(b), KVKK Art. 5(2)(c). We cannot run the service for you without it. |
| Comment authors | Comment text, username, link to the post, and what Signal derives from them | Legitimate interest — the account operator's, and ours in serving them, in reading and prioritising comments written to their own account: GDPR Art. 6(1)(f), KVKK Art. 5(2)(f). Under KVKK, also data made public by the person it is about: Art. 5(2)(d). |
Why legitimate interest is balanced for comment authors: the comment is public and was written to the account that receives it. Signal shows it only to that account's own team. It publishes nothing, builds no profile of the author, uses nothing for advertising, and sends nothing about the author to Telegram. Comment authors can object to this processing at any time; see § 7.
6. Data retention
- Account data: kept until you ask us to delete your account.
- Comments and mentions: kept while Signal is in use for the connected channel. Deactivating a channel or pausing an organization stops new collection and deletes nothing. Deleting a channel or an organization permanently deletes every comment and mention collected from it, at once. We also delete it on request — see § 7.
- Cloud Run access logs: retained per Google Cloud's defaults (typically 30 days).
7. Your rights
Depending on where you live (EU/UK GDPR, Türkiye KVKK, California CCPA), you may have the right to:
- Learn whether we process personal data about you, and request information about it
- Learn why we process it, and whether it is used for that purpose
- Know the third parties it has been sent to, in your country and abroad
- Access the personal data we hold about you, including what Signal derived from it
- Correct inaccurate data, and have third parties told about the correction
- Delete your data ("right to be forgotten"), and have third parties told about the deletion
- Export your data in a portable format
- Object to or restrict certain processing, including an outcome against you that results solely from automated analysis
- Claim compensation if unlawful processing causes you damage
To exercise any of these, email signal@getpeyk.com. We respond within 30 days. An access request is answered with everything we hold about you, including the AI classification, the priority score, and any reply Signal suggested.
If you are a comment author whose data was ingested because someone using Signal connected an account you commented on, you can ask us to delete that data by emailing the same address with the comment URL or your username on the source platform.
We delete it on your request, within 30 days, from Signal and from every copy we hold. One part is outside our control: comments are sent to Anthropic for classification, and Anthropic keeps API data for up to 30 days, or longer if its safety systems flag it.
If you are not satisfied with our answer, you can complain to the Personal Data Protection Board (Kişisel Verileri Koruma Kurulu, kvkk.gov.tr) or, in the EU, to the data protection authority where you live or work.
8. Children's privacy
Signal users must be at least 18 (see our Terms). Comment authors can be any age, and Signal has no way to know it. If you are a minor whose comment was collected, or the parent or guardian of one, you can ask us to delete it through § 7.
9. International data transfers
Where each party that receives data is located:
| Party | Location |
|---|---|
| Google Cloud (Cloud Run, Firestore) | Belgium, EU (europe-west1) |
| Google Firebase Authentication | United States. This cannot be configured to another region. |
| Anthropic | United States. Covered by the EU Standard Contractual Clauses and the EU-US Data Privacy Framework. |
| Telegram | Netherlands for accounts registered in the EEA. Telegram is operated by entities in the United Arab Emirates and the British Virgin Islands. |
10. Security
- All data is transmitted over TLS (HTTPS).
- Stored data is encrypted at rest by Google Cloud (AES-256).
- Access is gated by Firebase Authentication and role-based permissions inside each workspace.
- Credentials you provide — API keys and the access tokens for the accounts you connect — are stored in our database. They are protected by Google Cloud's encryption at rest, but not yet encrypted at the application layer. Until that ships, treat any credential in the system as one to rotate if it leaks elsewhere.
No system is perfectly secure. If you become aware of a vulnerability, please email signal@getpeyk.com.
11. Contact
Peyk Signal team
signal@getpeyk.com
Data controller: Nevcan Uludaş (see § 2)
12. Changes to this policy
We may update this policy from time to time. When we make material changes, we will notify you via email or a notice in the app at least 14 days before the change takes effect. The "Last updated" date at the top reflects the most recent revision.