Privacy Policy

Information We Collect

Depending on how you use Kanha.ai, we collect:

• Account, profile, and team information, including your name and email address.
• Site URLs and the public page content you select or authorize us to crawl.
• Generated questions and answers, datasets, training records, model artifacts, and related configuration.
• Prompts, service configuration, feature usage, quota usage, and other activity within the service.
• Information submitted for a trial, including contact details, work email, site URL, and sitemap URL.
• Billing and subscription identifiers. Card and other payment details are handled by the checkout provider and are not stored by Kanha.ai.
• Support requests, emails, and other communications you send to us.
• Technical logs such as IP address, browser and device information, request times, errors, and security events.
• Authentication cookies and preferences stored locally in your browser, such as theme settings.
• Public Device Benchmark summary fields when you choose to submit a result, such as the public result name or device label, scores, timing, model and runtime details, and device or browser summaries.

How We Use Information

We use this information to:

• Provide accounts, teams, crawling, dataset generation, training, model delivery, billing, and support.
• Operate trials and communicate about a trial or requested service.
• Measure usage, enforce plan limits, diagnose errors, secure the service, and prevent abuse.
• Improve product quality, reliability, and the experience of customers and website visitors.
• Meet legal obligations and respond to valid legal requests.

Crawling, Training, and Local Inference

Crawling selected pages, preparing datasets, and training models take place on infrastructure used by Kanha.ai and its service providers. The resulting model may answer locally in a visitor's browser. Browsers and devices differ, so Kanha.ai may adapt the runtime or delivery path, and performance can vary. When an answer is generated locally, the visitor's question and answer need not pass through a hosted inference API. Local inference does not mean that all setup, training, delivery, or service data stays on the visitor's device.

To deliver browser models without requiring each visitor to authenticate, Kanha.ai publishes trained browser model and runtime artifacts to publicly accessible Hugging Face repositories. Anyone with the repository or artifact URL can download those files. The artifacts are derived from the content submitted for training, so you should submit content only when you have the right to authorize this public delivery.

Service Providers and Sharing

We share information with providers that help operate Kanha.ai, only as needed for their work. These include Supabase for authentication and storage, Stripe or Razorpay depending on the checkout flow, Resend for email, and providers used for hosting, content delivery, crawling, and training. We may also disclose information when required by law, to protect people or the service, or as part of a business transaction subject to appropriate safeguards. We do not sell payment card details, which are handled by the checkout provider.

International Processing

Kanha.ai and its providers may process information in countries other than the one where you live. Privacy protections can differ across countries. We use provider agreements and other measures intended to protect information where applicable.

Retention and Deletion

We keep information only as long as needed to provide the service, maintain security and business records, resolve disputes, and meet legal obligations. Retention varies with the type of information and why it is held. You can request deletion, though we may keep limited records when required or reasonably necessary for these purposes. Removing a hosted artifact cannot recall copies that other people have already downloaded from a public repository.

Security

We use technical and organizational measures intended to protect information. No system, network, or storage method can be guaranteed completely secure.

Your Choices and Rights

You can update some account details in the service, control browser preferences, and choose whether to submit a public Device Benchmark result. You may ask to access, correct, delete, or receive a copy of your information, or object to or restrict some processing where applicable. These rights depend on where you live. We may need to verify your identity before completing a request.

People Under 18

Kanha.ai is not intended for people under 18. If you believe a person under 18 has provided personal information, contact us so we can review and remove it where appropriate.

Changes and Contact

We may update this policy as the service changes. We will post the updated policy and effective date here. For privacy questions or requests, email contact@kanha.ai.