Zatlantis

Privacy Policy

0. Operator Information

  • Operator (Individual): 長田慶汰朗
  • Address: 大阪府泉南郡熊取町青葉台2丁目20-６ (Japan)
  • Contact: support@zatsmin.com

1. Data Collection

Zatsmin collects the following data.

1.1 Information You Provide

  • Profile information (name, date of birth, gender, affiliation)
  • Age-gate information (18+ declaration and phone verification (SMS OTP) records)
  • Interest tags
  • Big Five self-report inputs
  • Update Log content (currently suspended; if relaunched, it will be included in this section)
  • Blog content
  • Structured Daily Survey data (including target user selection and nomination intensity)
  • Interaction checksheet responses and interaction feedback data
  • Text Upload input text (used temporarily only as needed for analysis processing)
  • Host menu selections and mël consumption facts
  • Profile visit (footprint) records
  • User-to-user direct message (DM) content (stored only as needed to provide chat functionality)

1.2 Special Handling of Gender Information

We collect gender information at account registration. However, because this service is not intended for dating between opposite sexes, collected gender information is never made public to other users and is never used as a filtering condition in recommendation search. Providing gender information is optional, and you may choose “Prefer not to say.”

We limit the purpose of collecting gender information to the following internal uses:

  • Detecting and correcting gender bias in personality-related calculations to improve algorithmic fairness
  • Using it as macro-level demographic data when creating anonymized or statistical information that cannot identify specific individuals

1.3 Automatically Collected Information

  • Device information (OS, browser)
  • IP address
  • Activity logs stored in Firestore
  • Ad display and interaction logs
  • Operational logs at host NFC tap (tap timestamp, host identifier, object identifier, etc.)
  • Profile view event logs (such as view timestamp)
  • Location information collected while the app is in use, only when location permission is granted

1.4 Payment Information

  • Payment information is processed via Stripe (we do not store your credit card details)
  • Payment-related metadata required for paid features, including Daily Survey reveal features

1.5 Face Photos (Raw Selfie Photos)

  • Raw selfie photos submitted at account registration are used temporarily to generate stylized profile images and for safety checks.
  • After processing, raw photos are deleted, and only minimum metadata is retained (such as selfieConsentAt, selfieConsentVersion, and selfieStatus).
  • Raw selfie image data is never disclosed or provided to other users or external partners.

1.6 Government ID Handling

  • At the initial stage, government ID submission is not required for either general user accounts or host accounts. Depending on feature nature, safety needs, legal requirements, or review requirements, we may request additional identity verification (including government ID submission).
  • Submitted ID images/PDF files are handled only for identity verification, incident response, fraud investigation, and duplicate account prevention.
  • ID numbers themselves are never included in public profiles or third-party data provision; hashed identifiers are used for duplicate checks.
  • Access to ID files and related metadata is restricted to personnel with minimum operational need and auditable pathways.

2. Purposes of Use

We use collected data for the following purposes:

  1. Providing the service
  2. Big Five trait calculation and analysis
  3. Mëlmin assignment
  4. Improving social recommendation algorithms
  5. Improving user experience
  6. Payment processing
  7. Creating anonymized/statistical data and providing it to third parties (see Section 3)
  8. In-app advertising display and related ad delivery/effectiveness measurement (mobile app)
  9. Issuance, management, and aggregation of in-app utility points (mël) and voting rights (Plus), and operation of community governance events (Fire Night)
  10. Calculation and update of public score (publicScore) and level (level)
  11. Operation of Daily Survey and interaction checksheets, and aggregation of related interaction data
  12. Delivery of host menu experiences (NFC tap, mël consumption, and usage history management)

When location information is used, we mainly use it for:

  • Estimating primary activity area
  • Improving precision of close relationship and recommendation candidate signals
  • Statistical analysis of interaction data derived from real-world relationship context

Repeated location sampling is based on low-frequency foreground-only collection (while the app is in use). This may generate derived signals such as estimated primary activity area and passive short-distance co-presence context with existing friends or accepted recommendation partners.

Raw latitude/longitude is not returned to other users as public profile data. For third-party provision and analysis (including BigQuery), we use only minimum derived features and aggregated statistics, and do not provide raw location history itself.

We do not present location data alone as definitive proof that users have physically met. Location data is used only as probabilistic context.

Also, we do not parse DM text in routine operations to infer plans to meet. DM handling does not assume semantic analysis of text except for purposes described in Section 3.6 (Trust & Safety and system operation).

3. External Provision and Handling of Anonymized Information

3.1 Principle of No Third-Party Provision Without Basis

Except as provided in this section and where required by law, we do not provide personal information (name, contact details, raw photo data, etc.) to third parties without prior user consent. The following cases are not treated as third-party provision of personal information:

  • Provision to payment providers such as Stripe within the scope necessary for payment processing (outsourcing)
  • Disclosure required by applicable law

3.2 Provision for Machine Learning and Research (Business Purpose of Zatsmin)

One of Zatsmin’s important business purposes is to visualize human relationships and contribute to next-generation technology and social research. For this purpose, we create anonymized or statistical information by applying irreversible processing to service-generated data (such as blog analysis outputs, interaction data, and survey-derived aggregated statistics) so that specific individuals cannot be identified, and we may continuously provide such data (including sales) to partner technology companies, research institutions, and other third parties.

3.3 Items Included in Provided Anonymized Information

Anonymized information provided to third parties may include the following. No personally identifying information such as names, raw selfies, raw DM chat text, or raw blog text is included (uploaded/submitted raw text is not provided or sold).

  • Big Five trait scores (derived from self-report and peer inputs)
  • Proprietary data based on interaction feedback
  • Age bands (e.g., 20s, 30s) and rough activity area (prefecture/state level)
  • In-service behavioral statistics (e.g., continuity trends in interaction features)
  • Statistical data of extracted linguistic features and abstract topic tendencies (non-PII metadata such as emotional tone and cognitive style)
  • Relationship-dynamics statistics from DM usage that do not depend on text semantics (reply intervals, length, rally density, etc.)
  • Statistical differential vectors between Public and Private blog posts
  • Aggregated interaction-checksheet UI operation statistics (screen dwell time, slider adjustments, etc.)
  • Daily Survey aggregated statistics (e.g., nomination-intensity distribution)

3.4 Method of Providing Anonymized Information

We provide the above data to third parties via encrypted and secure data linkage infrastructure (such as APIs and secure cloud storage).

3.5 User Opt-Out of Data Provision

If you do not want your data to be provided as anonymized information to third parties, you may opt out at any time from “Settings > Privacy Settings” in the app. Even if you opt out, basic service features (including Zaworld interactions) remain available.

3.6 Data Retention for Safety and System Reliability

As a principle, we do not access private communications or content between users beyond what is necessary to provide the service. To provide chat functionality (DM), message content is stored on servers for delivery and history display purposes. Retention periods follow Section 4.

In the following cases, we may exceptionally review or retain communication records temporarily:

  1. Trust & Safety (ensuring safety)
    For legal compliance, investigation/prevention of Terms violations (fraud, harassment, child abuse, etc.), or urgent protection of life/safety, we may temporarily review or retain communication records within the minimum necessary scope. For community safety and quality, we principally focus on metadata such as reply presence and conversation ending patterns. Routine operations do not perform AI semantic analysis of DM text; such analysis is limited to clearly necessary cases (e.g., legal obligations, life/safety protection, or severe Terms violations). Analysis outputs are retained only as anonymized statistics, and text data used for such analysis is discarded promptly after processing.
  2. System reliability (temporary buffer)
    To maintain stable delivery and functions (including asynchronous processing and reliable delivery), user-submitted data may be stored temporarily as electronic records on our communication systems (servers) until processing is complete. Temporary buffers are deleted promptly by system processing after completion.

4. Data Retention Periods

  • Account data, interaction data, and statistical data: retained until account deletion
  • DMs exchanged via Zaworld recommendations: if mutual friendship is not established, message body in the relevant thread is deleted within 30 days after recommendation expiration (expiresAt)
  • The above Zaworld DM deletion check is re-evaluated at each scheduled job run based on friendship status at that time. Even if users were friends when the grace period first elapsed, the thread becomes eligible for deletion in a later run if they are no longer friends.
  • DMs after mutual friendship is established, and DMs started from the Friends list: retained until explicit deletion or account deletion
  • Raw selfie photos: deleted after processing completes
  • Raw Text Upload text: deleted after analysis processing; only non-PII derived features are retained
  • Government ID files and identity-verification metadata: retained only as needed for verification, safety response, and legal requirements, then promptly deleted
  • Post-withdrawal data: fully deleted within 30 days after account deletion, unless retention is required by law or legitimate legal request

5. User Rights

Users have the following rights:

  • Access and correction of data
  • Data deletion request
  • Account deletion

For data portability inquiries, contact us at the address below. Whether and to what extent we can respond will be determined in light of applicable laws and our operational policies.

Contact: support@zatsmin.com

6. Security Measures

  • Access control through Firestore security rules
  • HTTPS communication
  • API protection (App Check enforcement by default, with managed exceptions only when required)
  • Secret management (Secret Manager usage and least-privilege principle)
  • Rate limiting, abuse prevention controls, and regular security audits

7. Machine Learning Data Processing

7.1 Blog Analysis / Weekly Summaries and Report Generation (Update Log Suspended)

  • We may use large language models, including OpenAI GPT-family models and successor models
  • We perform analysis required for Big Five traits, public-score-related derived features, and Mëlmin assignment
  • weeklySummaries is server-internal derived data and is not a weekly user push feature or always-visible home report
  • The Update Log feature is currently suspended; if relaunched, covered data scope will be updated under this policy

7.2 Interaction Checksheets, Daily Survey, and Interaction Feedback

  • Structured responses from interaction checksheets and Daily Survey are processed by the system
  • Aggregated statistics such as interaction feedback and nomination intensity are retained in anonymized form

7.3 Data Retention

  • Processing outputs are stored in Firestore and BigQuery as derived data for statistics/analysis
  • Raw text is not stored in BigQuery; only non-PII features and statistics are handled

8. Cookies and Tracking

  • When using the Web service, cookies may be used for session management and related purposes.
  • In the mobile app, we may display in-app ads (including interstitial ads) via Google AdMob. For ad delivery, fraud prevention, and effectiveness measurement, advertising identifiers (IDFA, GAID, etc.), device information, and usage information may be sent to Google LLC and its partners. Details are subject to each provider’s privacy policy.
  • Interstitial ads are controlled by policy-based intervals and excluded screens (e.g., login, profile setup, selfie verification, payment, and DM thread screens).
  • In the EEA, UK, and other applicable regions, we provide legally required consent/choice presentation through Google’s consent tool (User Messaging Platform, etc.). Personalization-related processing is controlled based on user choices.
  • On iOS, we may request tracking permission in accordance with App Tracking Transparency.
  • Some ad-related privacy choices may be reviewed or changed via screens linked from “Settings > Privacy Settings,” depending on region/platform requirements.

9. International Data Transfers

Data is stored in Asia-Pacific regions on Google Cloud Platform (GCP). In addition, due to service usage involving payments (Stripe), AI processing (OpenAI), and ad delivery (Google AdMob), data may be transferred to overseas locations managed by these processors/partners. We implement necessary contractual and organizational safeguards in accordance with applicable laws.

10. Contact

For privacy-related inquiries:

11. Changes to This Privacy Policy

This policy may be changed. For material changes, we will notify users via in-app notices, email, startup consent screens, or other methods we deem appropriate. We may adjust consent collection methods based on the nature of version updates.

Last updated: April 12, 2026