Artificial Intelligence Usage Policy (AI Policy)

Last updated: 18.02.2026 · Version: 1.0

This page explains how artificial intelligence (AI) and machine learning (ML)–based systems are used in the WIN application, which principles are followed, and users’ rights and obligations. The policy covers matching/recommendation systems, automated tools for security and moderation, and generative AI features.

Related documents

This policy should be read together with the following WIN documents:

Purpose and scope

Purpose

The purpose of this policy is:

• to explain which artificial intelligence and automated systems are used in WIN,
• to bring together the principles of transparency, authenticity, equity, and accountability in a single text,
• to ensure users are informed about data use, automated decisions, and rights to object, and
• to provide a framework consistent with national (Türkiye) and international (EU DSA, EU Artificial Intelligence Act, etc.) legislation.

Scope

What falls within the scope of the policy:

• Matching and recommendation systems (profile ranking, recommendations, visibility),
• Artificial intelligence for security and moderation purposes (fake profile, fraud, harassment, spam detection),
• Automated decisions (temporary restrictions, blocking transactions, account security measures),
• Optional artificial intelligence features (e.g. profile text suggestions, chat assistant),
• User content created or edited with generative AI (image, audio, text).

The rule regarding unauthorised interaction with the Services or Member Content using AI/ML tools developed by third parties is set out in Section 5 of the Terms of Use (Third-party applications); it should be read in a manner consistent with this policy.

Definitions

These definitions should be read consistently with Law No. 6698 on the Protection of Personal Data (KVKK), the EU General Data Protection Regulation (GDPR), and, where applicable, the EU Digital Services Act (DSA).

Areas of artificial intelligence use in WIN

Matching and recommendation systems

WIN may use recommendation systems to make the Services more useful. These systems may rank profiles shown to you, recommend them, or make certain profiles less visible. A summary of the data and signals used is set out in Section 8.1 (Matching and Recommendation Systems) of the Terms of Use; detailed data categories and legal bases are explained in the KVKK Information Notice and the Privacy Policy.

Recommendation systems are predictive in nature; a particular match or outcome is not guaranteed.

Security and abuse prevention

WIN may use AI-assisted tools for security and abuse prevention (e.g. detection of fake profiles, fraud, harassment). These tools may be used:

• to detect suspicious behaviour patterns,
• to evaluate bot, spam, or fake account signals, and
• to support human moderators in content moderation (prohibited content, hate speech, suspicion of minors, etc.).

Signals produced by AI-assisted mechanisms are assessed together with human review where appropriate. This approach is consistent with Section 8.3 (AI / Artificial Intelligence Use) and Section 8.4 (Automated Decisions) of the Terms of Use.

Community dynamics and launch phase

During the platform’s launch and initial stages (“the first few months”), WIN may use a limited number of system-managed accounts (bots) so that users do not experience an empty app and to encourage community interaction. These accounts may remain active for a temporary period to maintain platform fill and interaction quality.

WIN may offer optional artificial intelligence features as part of providing the Services (e.g. profile text suggestions, conversation starters). Such features:

• are clearly labelled when rolled out and users are informed,
• are not mandatory; they are left entirely to user choice, and
• have their data processing explained in advance under the Privacy Policy and KVKK Information Notice.

As such features are offered, short explanations on the relevant screens and, where necessary, a separate consent/approval mechanism may be provided.

Content created with generative AI

WIN is a dating / matching application. In this context, trust and authenticity are fundamental; artificial intelligence must never be a deceptive element. WIN has adopted a specific policy on the use of content created or edited with generative AI. The rules below apply both now and when such features are offered in the future.

Prohibitions (what not to do)

The following acts are considered breaches of the Terms of Use and Community Guidelines:

• Misleading photo / image use: Using images that are not yours or that have been substantially altered with AI (age, body, face, gender expression, etc.) in your profile as if they were yours.
• Pretending to be another person: Using another real person’s likeness (deepfake, face swap, AI portrait, etc.) on your profile or causing yourself to be recognised as that person.
• Presenting AI content as real: Using AI-generated or edited image, audio, or video without disclosure or in a way that gives the impression of being “real / mine” (failing to use an “Created with AI” label when WIN requires it may also be assessed under this heading).
• Serious deviation in age or identity: Using AI or other tools to materially mislead as to your appearance, age, or identity (e.g. presenting a minor as an adult or vice versa).
• Fake identity in chat / messages: Using AI-generated text, audio, or imagery in messaging as if you were someone else or as if it belonged to a real person (e.g. fake voice message, fake photo sent as if genuine).

This list is not exhaustive; deceptive or misleading AI use of a similar nature may also constitute a breach. Detailed prohibited content and behaviour are set out in Section 4 (Content rules) and Section 4.1.2 (Profile photo and visual standards) of the Community Guidelines; deepfakes / AI-generated portraits are specifically addressed.

Recommendations (what to follow if you use AI)

WIN may allow limited and transparent use of AI-created or edited content (now or in the future). In that case, you are advised to comply with the following, and (when offered as a product feature) compliance may be made mandatory:

• Disclose: Clearly state on your profile or in the relevant content that AI was used. If WIN provides an “Created with AI” label or a similar field, use it.
• Represent yourself honestly: Edits made with AI (e.g. background, light filter) must not materially mislead as to your real appearance or identity. The aim is that the other party’s expectations remain reasonable when meeting you in person.
• Use only in permitted areas: Uploading content produced with external tools in areas where WIN does not offer an AI feature (e.g. profile photo) that is misleading may fall within the prohibitions.
• Do not use others’ content without permission: Processing another person’s photo or image with AI and using it on your profile is prohibited; this infringes both copyright and identity (impersonation).

These recommendations are the practical expression of the authenticity principle; updates will be published through in-app notices and this policy.

Consequences of breach (sanctions)

Where misleading or prohibited AI content use is detected or assessed following a complaint, WIN may apply one or more of the sanctions below. Outcomes depend on the severity, repetition, and intent of the breach (consistent with Section 7 of the Terms of Use and the sanction ladder in the Community Guidelines):

• Complaints and detection: Other users may report profiles they believe use misleading AI content via Report. WIN assesses complaints and automated detection tools; human review is performed where necessary.
• Objection: Objections to account restrictions or closure are made through Section 7 of the Terms of Use and the in-app objection/support channel.

When generative AI features are offered, this policy and in-app copy will be updated; prohibitions and consequences apply under misleading content and profile accuracy rules whether or not an AI feature is offered.

Principles

WIN adheres to the following principles in its use of artificial intelligence.

Transparency

• We explain what we do: The purposes for which AI is used (matching, security, optional features) are summarised in this policy and Section 8 of the Terms of Use.
• Data use: Which data categories are used in AI/recommendation systems is stated in the Privacy Policy and KVKK Information Notice.
• Automated decisions: Where automated decisions directly affect you, information is provided to the extent possible and a channel for objection/review is offered (Terms of Use Sections 8.4, 8.5).

For users in the EU, algorithmic transparency and content moderation information under the DSA is additionally explained on the DSA Compliance Page. The DSA, in particular, envisages information in the terms and conditions on algorithmic decision-making and human review in content moderation (consistent with Regulation (EU) 2022/2065).

Authenticity

• Honest representation: AI must not prevent users from presenting themselves in an honest and authentic manner. WIN aims to foster meaningful and reliable connections.
• Misleading AI content: Using AI-generated or edited content to present yourself or your intent in a misleading way is not acceptable; breaches are subject to sanctions under the Community Guidelines and Terms of Use.

Equity

• Non-discrimination: AI systems are designed and operated so as not to cause unfair discrimination on grounds such as race, ethnic origin, gender, sexual orientation, religion, disability, etc.
• Bias mitigation: When new AI features are launched, possible impacts on different user groups are assessed; identified risks are sought to be reduced.
• Equal access: The recommendation and matching experience is considered to be as fair and consistent as possible; however, perfect equality cannot be guaranteed in every case due to technical limitations.

These principles should be read consistently with lawful and fair processing of personal data and the prohibition of discrimination (Art. 4) foreseen in KVKK No. 6698.

Accountability

• Decisions and processes: Automated decisions are intended to be supported by human review where appropriate, and an objection route is offered.
• Transparency reports: Periodic Transparency Reports are published on content moderation and security processes; these reports may be updated to cover algorithmic and human moderation.

Security and integrity

• Abuse prevention: AI is used to enhance community safety against risks such as fake accounts, fraud, harassment, and spam.
• Data security: Data used for AI systems is protected under the Privacy Policy and technical security measures.

Privacy

• Data minimisation: Only necessary and proportionate data is used for AI; purposes are explained in the KVKK Information Notice and Privacy Policy.
• Sale to third parties: WIN does not sell your personal data for money; transfers of data used for AI to third parties are subject to the transfer rules in those documents.

Data and privacy

Data categories, processing purposes, retention periods, and legal bases used for AI and recommendation systems are set out in detail in:

• Privacy Policy
• KVKK Information Notice

In cases of profiling and automated decision-making, your rights to information and to object / request human intervention under KVKK Art. 11 and (for users in the EU) GDPR Art. 22 are also set out in those documents and summarised in the “Automated decisions and objection” section below.

Automated decisions and objection

WIN may apply automated decisions in some cases for service security and fraud prevention. Examples: temporary restriction of certain features in suspicious behaviour patterns, blocking a transaction in suspicious payment/transaction attempts, triggering account security steps on bot/spam detection (Terms of Use Section 8.4).

• Information: If such an automated decision has a significant effect on you, information is provided to the extent possible.
• Objection / review: A channel is provided to object or request human review. The objection process and timeframes are explained in Section 7 (Moderation, Suspension, Termination) of the Terms of Use and, where applicable, on the DSA Compliance Page. Where a written structured application is required, the Objection Form may be used.
• KVKK / GDPR: Your rights to object to automated decisions and request human intervention are also set out in the KVKK Information Notice and Privacy Policy (KVKK Art. 11, GDPR Art. 22).

User obligations

• Misleading AI content: Using AI-generated content (image, audio, video, text) to misrepresent yourself, your appearance, or your intent constitutes a breach of the Terms of Use and Community Guidelines. A breach may result in content removal, warning, feature restriction, temporary suspension, or permanent closure of the account (for details, see the prohibitions and consequences of breach in the Content created with generative AI section above).
• Third-party AI/ML tools: You undertake not to use or develop third-party applications that interact with the Services or Member Content without WIN’s written permission (including artificial intelligence, machine learning, or similar automation) (Terms of Use Section 5).
• Complaints: If you consider another user’s AI use to be a breach, you may use the in-app Report function or the relevant notification channel such as the Content Infringement Form or Complaint and Notification Form (Community Guidelines, DSA Compliance Page).

EU and international framework

Digital Services Act (DSA)

Where services are provided in the EU, under Regulation (EU) 2022/2065 (Digital Services Act):

• information in the terms and conditions on content moderation and algorithmic decision-making / human review,
• transparency regarding recommendation systems and ranking, and
• statement of reasons and objection/complaint mechanisms

are additionally explained on the DSA Compliance Page. WIN’s legislative index includes a reference to the DSA and the path to the relevant document (WIN Regulation index — Digital Services Act).

EU Artificial Intelligence Act (EU AI Act)

The EU Artificial Intelligence Act (Regulation (EU) 2024/1689 — AI Act) introduces a risk-based framework (unacceptable risk, high risk, limited risk, minimal risk). The AI systems used in WIN (matching/recommendation, security/moderation) are generally of a nature that may be assessed in the limited risk or minimal risk category; high-risk AI use is not targeted. Because the Act’s entry-into-force dates and additional obligations may change, this policy and related pages will be updated as services are offered in the EU market.

Turkish legislation

• KVKK No. 6698: Processing of personal data, profiling, and automated decision-making are consistent with the KVKK Information Notice and Privacy Policy.
• Law No. 5651: Hosting provider liability and traffic data retention are assessed together with content moderation processes in the context of the Terms of Use and Transparency Reports.

Controls and preferences

As a user, you may:

• change your preferences and filters in settings (age, distance, matching preferences, etc.),
• manage your location permission from device settings (Privacy Policy — location data), and
• block or report unwanted users,

and you may expect these preferences to feed into the recommendation systems. How recommendation systems work and any adjustable options are summarised in Section 8.5 (Controls and Transparency) of the Terms of Use.

Contact and feedback

• General support and account: In-app Help or support@whoisnextapp.com.
• KVKK / data rights: Data controller contact details and application routes in the KVKK Information Notice.
• Automated decision / objection: Objections regarding account restrictions or termination are made through Section 7 of the Terms of Use and the in-app/objection form or support channel.
• Feedback on the AI policy: If you have questions or feedback on this policy or AI use in WIN, you may contact us through the general support channel (support@whoisnextapp.com).