PRIVACY POLICY

Let’s Whisper s.r.o. (Yasan)

Effective Date: 19 October 2026
Last Updated: 19 October 2026
Version: 2.0 (replaces the Privacy Policy effective 1 January 2026)


1. WHO WE ARE

This Privacy Policy explains how Let’s Whisper s.r.o., Křižíkova 148/34, Karlín, 186 00 Praha 8, Czech Republic, Company ID 23849959, registered in the Commercial Register maintained by the Municipal Court in Prague, Section C, Insert 433436, operating under the brand Yasan (“Yasan”, “we”, “us”), handles personal data.

Contact for all privacy matters: [email protected] · Data box (datová schránka): 3e7q2vh · Postal address as above.

We have not appointed a data protection officer, as we are not required to. We are established in the Czech Republic; our lead supervisory authority is the Czech Office for Personal Data Protection (Úřad pro ochranu osobních údajů, www.uoou.gov.cz).

United Kingdom. We have no establishment in the UK. Where UK GDPR applies to our processing, this Policy applies accordingly, references to the GDPR include the UK GDPR and the Data Protection Act 2018, and the competent supervisory authority is the Information Commissioner’s Office (ico.org.uk). We will designate a representative in the UK under Article 27 UK GDPR when our UK processing requires it, and will publish their details here.


2. SCOPE — WHO THIS POLICY IS FOR

This Policy applies to:

  • Website visitors of yasan.ai, letswhisper.ai and our other websites;
  • Prospects — people at businesses who contact us about our services (demo requests, events, trials);
  • Customers and their Users — businesses that use the Yasan platform (the “Service”) and the individuals who use it on their behalf;
  • End Users — shoppers and visitors who interact with the Yasan chat, search or other features on a Customer’s website, or who write to a Customer whose mailbox is connected to the Service. For End Users we are almost always a processor acting for the Customer; see Section 6 and our End User Notice at https://yasan.ai/ai-assistant-privacy.

3. OUR TWO ROLES — CONTROLLER AND PROCESSOR

We act as controller for We act as processor for
Account, billing and usage data of Customers and Users; website visitor data; prospect and marketing data; support communications; security logs of the Service, including the limited technical data of End Users described in Section 6.4. We decide why and how this data is processed, and this Policy describes it. Everything the Customer brings into the Service or that the Service collects for the Customer: catalogue data, connected emails and systems, chat conversations, search queries, analytics events, End User contact and order data (“Service Content”). The Customer is the controller; we process on its documented instructions under our Data Processing Agreement (https://yasan.ai/dpa).

4. DATA WE PROCESS AS A CONTROLLER

Category Data Purpose Legal basis
Account & billing data Name, work email, phone, job title, company name and ID, billing address, VAT ID, payment status and transaction history (card details are held by our payment processor Stripe, not by us), Platform store identifiers (e.g. Shoptet/Shopify shop ID) Creating and administering accounts; invoicing and collection; fraud prevention; tax and accounting records Art. 6(1)(b) contract; Art. 6(1)(c) legal obligation (accounting, tax); Art. 6(1)(f) legitimate interest (fraud prevention)
Usage & telemetry IP address, device and browser information, timestamps, pages and features used in the administration interface, configuration changes, error and performance logs, session recordings of the administration interface (Smartlook, Sentry) Securing and operating the Service; diagnosing errors; understanding how the product is used; improving it Art. 6(1)(f) legitimate interest (security, product improvement); consent where required for analytics cookies
Communication data Messages to our support, sales and in-app chat; demo requests; feedback; call notes Support; pre-contractual steps; improving the Service Art. 6(1)(b) contract and pre-contractual steps; Art. 6(1)(f) legitimate interest
Customer communications Work e-mail and name of Customers’ Users, interaction with our e-mails Product updates, release notes, service notices, onboarding tips and occasional newsletters about the Service to existing Customers and trial users Art. 6(1)(f) legitimate interest in informing our Customers about the Service they use (service notices are sent under Art. 6(1)(b)). You can unsubscribe from non-essential updates at any time via the link in each e-mail.
Website visitor data IP address, device and browser data, pages viewed, referrer, cookies and similar technologies (see Section 11) Running and securing our websites; measuring audience; marketing where consented Art. 6(1)(f) legitimate interest (operation, security, essential measurement); Art. 6(1)(a) consent for non-essential cookies
Partner & vendor contacts Business contact details of suppliers, partners, Platform contacts, investors and advisers Managing business relationships Art. 6(1)(b) contract; Art. 6(1)(f) legitimate interest
End User security data IP address and request metadata of End Users of our widgets, used transiently for rate limiting and bot protection; error and performance logs (Sentry), deleted within 90 days; short-lived security logs Protecting the Service and all Customers against abuse and attacks; diagnosing errors Art. 6(1)(f) legitimate interest in security and integrity of the Service (see Section 6.4)

Automated decision-making. We do not take decisions producing legal or similarly significant effects on individuals based solely on automated processing.

Where the data comes from and is it required. Providing account and billing data is necessary to conclude and perform the contract; without it we cannot provide the Service. All other data is voluntary or collected automatically as described.


5. SERVICE CONTENT — DATA WE PROCESS AS A PROCESSOR

When a Customer uses the Service, we process the following on the Customer’s behalf and instructions, as its processor:

  • Catalogue and knowledge data: products, prices, stock, categories, content pages, FAQs, instructions and rules the Customer gives the assistant.
  • Data from the Customer’s e-commerce platform (Shoptet, Shopify, WooCommerce or a custom API). Where a Customer connects Shopify, we access through the Shopify API, under the scopes shown at installation: the product catalogue (titles, descriptions, variants, prices, stock, images, translations); orders and their line items, notes and comments, fulfilment, financial and return status, and the customer on those orders (name, e-mail, phone number, billing and shipping address); the published theme’s settings (to check whether the widget is enabled); and the store’s languages and markets (so that answers use the right language and currency). Orders — including the customer’s contact details on them — are read on demand to answer a specific question (for example “where is my order”) or perform an action the merchant has allowed (for example a shipping-address change, which may include updating the phone number); we do not keep a copy of the store’s customer list. The same principles apply to other platforms, within the data each platform’s API exposes.
  • Chat Widget data: conversations between the Customer’s website visitors and the assistant (messages, AI responses, session identifier, language, page URL), e-mail addresses that visitors provide for follow-up or order verification (verification codes are stored only as hashes, unless the visitor types the code into the chat), order data retrieved from the Customer’s systems after verification, and shopping-cart context used to answer the question.
  • AI Search data: search queries that visitors submit, the products shown and the assistant’s answer, follow-up messages, the filters applied (category, brand, parameter, a price limit written in the query) and which results were on screen, page address (origin and path only — query strings and fragments are discarded before storage), a basket summary (item count, total value, currency), language, timestamps and technical delivery data. As-you-type suggestions are served from the catalogue and are not stored; the query and its context (for example the candidate products, filters and earlier queries in the same shop) may be sent to AI models that assess relevance and help find suitable products — without any visitor or session identifier — of which only technical results without the query text are kept, for a limited time. A daily sample of the previous day’s queries is re-run through the same relevance models to measure search quality for that shop; the sample stores counts only, not query text, and is deleted with the underlying search record.
  • Email Agent data: e-mails, attachments, headers, labels and folders in the Customer’s connected mailbox or help desk; drafts and replies; data looked up in connected systems (orders, invoices, delivery status, customer profiles); actions executed.
  • Analytics data: events of the Customer’s website visitors — linked to a random pseudonymous visitor identifier generated in the browser only where the visitor has consented — and statistics derived from them, exclusively for that Customer.
  • Users’ content: notes, learnings, feedback and configuration created by the Customer’s Users.

Principles that apply to all Service Content:

  • Per-Customer isolation. Service Content is used exclusively to provide the Service to the Customer it originates from. We do not pool or cross-reference conversations, searches or e-mails across Customers, and we do not build profiles of individuals across websites.
  • No training of general models. We do not use Service Content to train, fine-tune or improve general-purpose AI models — neither ours nor those of our model providers. Our contracts with model providers prohibit training on and retention of the data beyond processing the request (subject to short abuse-monitoring retention where the provider requires it). We do use Service Content to configure and tune the assistant for the particular Customer (e.g. prompt tuning, retrieval logic, learned rules) and to create aggregated, anonymised statistics.
  • Limited human access. Our staff access Service Content only where necessary for support requested by the Customer, for security investigations, for quality review of the assistant’s behaviour for that Customer, or where required by law, and always under confidentiality obligations.
  • Platform privacy requests (Shopify). When Shopify sends us a customers/data_request, customers/redact or shop/redact webhook, we log the request with a 30-day deadline and, within that period, either provide the merchant with an export of the personal data we hold about that customer or permanently remove the personal data concerned — including message bodies, attachments and derived embeddings — from all our systems (records of actions carried out in the merchant’s store, such as an address change, stay with the merchant’s support history), subject only to backups overwritten in the ordinary cycle and to legal retention duties. Comparable requests from other platforms and direct requests from merchants are handled within the same 30-day period.
  • Google API Services. Our use of data received from Google APIs (e.g. Gmail) complies with the Google API Services User Data Policy, including the Limited Use requirements: we use such data only to provide and improve user-facing features of the Service, do not transfer it except to sub-processors necessary for the Service, for security or legal reasons, or as part of a merger or acquisition, do not use it for advertising, and do not allow humans to read it except as described above or with the Customer’s explicit consent.

6. END USERS — IF YOU USED A Yasan CHAT OR SEARCH ON A SHOP’S WEBSITE

6.1. Who is responsible. The operator of the website or shop you were using is the controller of your personal data. Yasan is its processor and processes your data only on the operator’s instructions. A plain-language explanation of what is processed, why, for how long and what your rights are is published in our End User Notice at https://yasan.ai/ai-assistant-privacy (Czech, English, German). The operator’s own privacy policy applies in addition.

6.2. You are interacting with AI. The chat and search assistants are AI systems. This is disclosed directly in the widget, as required by Article 50 of the EU AI Act. AI answers may be inaccurate; for anything binding, rely on confirmation from the shop. The assistant does not take decisions with legal or similarly significant effects on you.

6.3. Consent and identifiers. The widgets store only strictly necessary items in your browser (such as a session cookie, the conversation state and your consent choice). A random visitor identifier that links your visits is created only if you grant analytics consent in the shop’s cookie banner — in the UK in the same way as in the EU — and is deleted from your browser as soon as consent is withdrawn. We never use these identifiers for advertising and never share them with advertising networks.

6.4. What we process as a controller about End Users. Only what is needed to keep the Service secure for everyone: your IP address and request metadata, used transiently for rate limiting, abuse prevention and bot protection (Cloudflare Turnstile) and recorded in the error and performance logs of the Service (Sentry), which are deleted within 90 days, and short-lived security logs retained for no longer than 12 months. A daily-rotating pseudonymous value derived from IP address and browser details is used to count visitors; its key changes every calendar day and records are not linked across days. Legal basis: Article 6(1)(f) GDPR — our legitimate interest in the security and integrity of the Service. We do not use this data for profiling or marketing.

6.5. Retention. Conversations, search records and analytics events are deleted no later than 24 months after they were created, and within 30 days after the shop’s Yasan account is closed, subject to backup cycles. Records which cannot be attributed to any individual (for example searches made without consent and outside a conversation) cannot be retrieved for a data-subject request and are deleted within the same period.

6.6. Your rights. Please address requests to access, correct, delete or restrict your data to the operator of the shop where you used the assistant. If you contact us at [email protected], we will forward your request to the operator without undue delay and assist them; we will not answer on the merits on their behalf. For the security processing in Section 6.4, where we are the controller, you may exercise your rights with us directly.


7. ARTIFICIAL INTELLIGENCE AND MODEL PROVIDERS

The Service uses LLMs and other AI models from third-party providers through enterprise APIs — currently Google (Gemini), OpenAI, Anthropic and Amazon Web Services (Amazon Bedrock) for generating answers and finding products, and TypeSafe AI (with OpenAI as fallback) for assessing the relevance of search results. Depending on the feature, the text of a query, message or e-mail, the relevant product information and page context are sent to the model provider to generate an answer; the relevance-assessment providers receive the query and its context (for example the candidate products and filters), without any visitor or session identifier. Our agreements with all of these providers prohibit the use of this data for training their models. Google, OpenAI, Anthropic and Amazon Web Services retain request data only as long as needed to process the request and to prevent abuse; TypeSafe AI processes requests under a zero-data-retention arrangement: it does not store the content of requests or responses after processing them, keeps only technical logs and usage metrics that do not contain that content, and receives no visitor or session identifiers from us. TypeSafe AI is not certified under the EU–U.S. Data Privacy Framework; transfers to it rely on Standard Contractual Clauses (with the UK Addendum). We may change or add providers; the current list is at https://yasan.ai/subprocessors.


8. SHARING OF DATA

We do not sell personal data. We share personal data only with:

  • Sub-processors that help us operate the Service (hosting and infrastructure, AI model providers, analytics, transactional e-mail, bot protection, support and engineering tools). Each is bound by a data processing agreement. The current list, with location and transfer mechanism, is published at https://yasan.ai/subprocessors.
  • Platforms through which you installed the Service (e.g. Shoptet, Shopify, WooCommerce), to the extent required for installation, billing, app-store operation and compliance with the Platform’s rules.
  • Payment and accounting providers (Stripe, our accountants and auditors), banks and tax authorities.
  • E-mail marketing provider (Brevo) for product updates and newsletters to Customers’ Users who have not unsubscribed.
  • Professional advisers, insurers and, where required, public authorities, courts or law-enforcement bodies.
  • A successor in the event of a merger, acquisition, financing or sale of assets, who must honour this Policy.

9. INTERNATIONAL TRANSFERS

We are based in the EU and host the Service primarily in the EU. Some sub-processors (notably AI model providers, bot protection and e-mail delivery) process data in the United States or may route processing through other regions.

  • From the EU/EEA: transfers to the US rely on the EU–U.S. Data Privacy Framework where the recipient is certified and, otherwise or in addition, on the European Commission’s Standard Contractual Clauses with supplementary measures where needed. Transfers to other third countries rely on an adequacy decision or on Standard Contractual Clauses.
  • From the UK: transfers rely on the UK Extension to the EU–U.S. Data Privacy Framework (UK–US “data bridge”) where the recipient is certified, and otherwise on the UK International Data Transfer Addendum to the EU Standard Contractual Clauses (or the IDTA). Transfers from the UK to the EU are covered by the UK’s adequacy regulations for the EEA.

A copy of the relevant safeguards can be requested at [email protected].


10. RETENTION

Data Retention
Account and billing data Duration of the relationship, then 10 years for accounting and tax records (Czech law); other account data deleted within 30 days after account closure
Service Content (Customer’s processor data) Chat and search conversations and analytics records: removed no later than 24 months after creation. E-mail support history (threads, drafts, replies): kept for as long as the Customer’s account exists, because agents need past conversations, and deleted earlier on the Customer’s request or on a platform privacy request. Order and customer data from connected platforms: read on demand and not stored separately (order details quoted in a conversation are kept with that conversation); no copy of the customer list is kept. Integration credentials and access tokens: erased immediately when the integration is disconnected or the app is uninstalled. Everything deleted within 30 days after the Customer’s account is closed, on the Customer’s request or by us under the Terms (backups overwritten in the ordinary cycle), unless law requires retention. Ending a paid plan or trial does not by itself close the account
Trial accounts Kept while the account exists; may be deleted 30 days after the trial ends without subscription (Terms 3.6)
Usage and security logs 12 months
Customer communication preferences Until you unsubscribe or 3 years after the end of the relationship
Support communications 3 years after the ticket is closed
Contractual and dispute records For the limitation period, generally up to 10 years

11. COOKIES AND SIMILAR TECHNOLOGIES ON OUR WEBSITES

On yasan.ai and our other websites we use strictly necessary cookies (security, session, consent choice) without consent, and analytics, session-recording and marketing cookies only with your consent given in our cookie banner, which you can change at any time via the cookie settings link in the footer. A detailed list of cookies, providers and lifetimes is available in the banner’s “details” view.

Widgets on Customers’ websites are different: they store only strictly necessary items in the End User’s browser, and they create a visitor identifier only with consent (or where a lawful exemption applies). The full list of storage items is in the End User Notice. The website operator is the controller.


12. SECURITY

We implement technical and organisational measures appropriate to the risk, including encryption in transit (TLS 1.2+) and at rest (AES-256), role-based access control and multi-factor authentication for production access, logging, vulnerability management, segregated environments, encrypted backups, and contractual and confidentiality obligations for all staff and sub-processors. Details are in Schedule 2 of our DPA.


13. YOUR RIGHTS

Where we are the controller, you have the right to: access your personal data; rectify inaccurate data; erase data (subject to our legal retention duties); restrict processing; data portability; object to processing based on legitimate interests, including direct marketing (which we will stop immediately); withdraw consent at any time without affecting prior processing; and not be subject to solely automated decisions with legal effects.

How to exercise them: e-mail [email protected] from the address associated with your account, or write to our registered office. We will respond within one month (extendable by two months for complex requests, in which case we will tell you). We may ask you to verify your identity. Requests are free unless manifestly unfounded or excessive.

Complaints. You can complain to us first — we take complaints seriously and will acknowledge within 30 days and respond without undue delay (for UK individuals this is our complaints procedure under section 164A of the Data Protection Act 2018; a complaint form is available on request at [email protected]). You also have the right to lodge a complaint with a supervisory authority, in particular the Czech Office for Personal Data Protection (Pplk. Sochora 27, 170 00 Praha 7, www.uoou.gov.cz), the authority of your habitual residence or place of work in the EU/EEA, or the UK Information Commissioner’s Office (ico.org.uk) if you are in the UK.

End Users: see Section 6.6.


14. CHILDREN

Our websites and the Service are intended for businesses and adults. We do not knowingly collect data from children. The chat and search assistants on Customers’ websites are not intended for children below the age of digital consent in their country (15 in the Czech Republic, 16 in Germany, 13 in the UK). We do not verify the age of users of these assistants; if we learn that we hold data of such a child, we will delete it or help the Customer delete it.


15. CHANGES TO THIS POLICY

We may update this Policy. The current version is always at https://yasan.ai/privacy with the date of the last update. For material changes we will notify Customers by e-mail or in-app notice before the change takes effect.


Let’s Whisper s.r.o. · Křižíkova 148/34, 186 00 Praha 8, Czech Republic · IČO 23849959 · [email protected] · Data box 3e7q2vh

  • Terms of Service
  • Privacy Policy
  • Data Processing Agreement
  • Sub-processors
  • Shopify app terms
  • Shoptet add-on terms
  • AI assistant privacy notice

Previous versions ·

Let’s Whisper s.r.o. · Křižíkova 148/34, 186 00 Praha 8 · IČO 23849959 · [email protected]