Privacy Policy
Last updated 2026-11-18
This policy explains what Laydyne ("we", "us") collects when you use laydyne.com and the Laydyne Studio, MCP and API (the "Service"), why we collect it, who processes it on our behalf, and the choices you have. We handle personal information in accordance with applicable data-protection law.
1. What we collect
We collect only what the Service needs to work:
- Account data. Your email address, a hashed password (never the password itself), the time you signed up and the language you chose. Email confirmation and password-reset codes are kept only until they expire.
- Session data. A session cookie that identifies your login, the time it was created, and the IP address and browser user-agent string at that time. Sessions expire after 30 days at the latest.
- Projects you save. Your working space normally stays in your browser. A cloud project that you save on a paid plan contains the spatial model, the recent conversation and any images you added as floor-plan underlays. Saving is manual, not automatic.
- AI workspaces. When an AI app connects with sign-in (OAuth) and you approve it, the project it edits is saved on our servers as an AI workspace, on every plan including Free, with recent versions kept for undo and conflict recovery. You can open or delete it in Studio under Connect AI.
- Viewing links. When you create a viewing link, we store a copy of the project as it is at that moment, a 2D plan of its floor area, the name and note you give, the expiry date and how many times the link was opened. Anyone who has the link can open the copy without an account; the link is not listed or indexed. The copy is deleted when you stop the link or when it expires (after at most 30 days), and the remaining record a week later.
- AI requests. When you use the built-in assistant or AI-assisted drafting, the conversation, the current scene and the attachment you provided are sent to the selected AI provider. We record which model was used, token counts and an estimated cost so we can apply plan limits; attachments sent only for AI processing are not stored separately by us.
- MCP relay data. For agent links, and for tools that need an open Studio tab (pictures, 3D views, PDF), tool requests and results between your AI agent and Studio pass through our servers and are held briefly for delivery. Agents that call tools inside the page (WebMCP) do not use that relay.
- MCP connection records. When an AI app works with your account through a connection link or sign-in (OAuth), we count by day which kind of app it was (for example Codex or ChatGPT, with the short name the app reports for itself), how many connections and tool calls it made and how many failed, and when you created a link or approved or declined a connection. We do not keep the tool requests or results. Before sign-in, connection attempts are counted only by step, a fixed reason when a step fails, kind of app, major/minor client version and a fixed user-agent family: the network address and browser user-agent are used in memory to estimate distinct sources per day and over a seven-day window and are not stored; only aggregate counts and sketches are kept for seven days.
- Billing data. Payments are handled by Stripe. We store the Stripe customer and subscription identifiers, the plan, the billing interval, the paid period and the subscription status. We never receive or store card numbers.
- Diagnostics. Fixed error codes, counts and timestamps from our servers and from browser error reports, without message bodies, email addresses or credentials. These records are kept for a limited period, currently seven days.
- Usage analytics. On laydyne.com we use Google Analytics to count page views and a small set of product events (such as opening Studio, starting a sign-up, exporting a project or clicking Upgrade). It sends the pages you view, the event names, your browser and device type, a cookie identifier and a truncated IP address to Google LLC in the United States, for the purpose of measuring how the site is used. It is loaded only on the production site.
- First-party usage counts. Independently of Google Analytics, we count page requests, page views and feature use, without setting analytics cookies or storing personal identifiers, and keep only daily totals and anonymous visitor-count estimates for up to 400 days. These counts are kept by public page path, page group, the referring site’s host name or a fixed referrer category (such as search, AI assistants, direct and internal), the country code Cloudflare reports for the request, device type, browser family or the name a bot gives for itself, and whether the request looked like a person’s browser, a self-declared bot or other automated software; event payloads contain no private paths, query strings or full referrer URLs. That class is worked out in memory from the request’s headers and the network it came from. Network addresses are used transiently to limit event requests; raw addresses are not written to analytics logs or storage. Addresses and browser user-agent strings are also used only in memory to estimate daily visitors. The site’s analytics opt-out also stops these browser events.
- Bot protection. Sign-in and sign-up forms are protected by Cloudflare Turnstile, which may process your IP address and browser signals to tell people from bots.
2. Why we use it
- To create and secure your account, deliver confirmation and reset emails and keep you signed in.
- To provide the features you ask for: building and saving spaces, running simulations, relaying MCP tool calls and answering AI requests.
- To bill paid subscriptions, confirm paid periods and apply plan limits such as monthly AI model calls and storage.
- To keep the Service reliable and secure: detecting abuse, diagnosing failures and preventing automated attacks.
- To understand how the public site is used so we can improve it, in aggregate.
- To respond when you contact us, and to meet legal obligations.
3. Who processes it for us
We do not sell personal information. The following providers process data on our behalf, each limited to the purpose listed:
- Cloudflare, Inc. hosts the Service (Workers, D1 database, Durable Objects), provides Turnstile bot protection and delivers sign-in, confirmation and password-reset emails through Cloudflare Email Sending. Data may be processed in Cloudflare data centres in several countries.
- Stripe, Inc. and Link. Pro subscriptions are sold by Link, Stripe’s merchant-of-record service, which receives the details you enter at checkout as an independent controller under its own privacy policy, sends receipts and billing emails, and handles payment support. We receive only the identifiers and subscription status listed above.
- OpenAI and Google (Gemini) receive the content of AI requests when you use AI features. We send only what a request needs, through their business APIs. Under the API terms that apply to us as of the date above, these providers do not use API inputs to train their models and may retain them for a limited period for abuse monitoring. We do not control their terms. For an ambience image (Pro and Max), the picture of your 3D view and any note you add go to OpenAI; the generated picture comes back to your browser and we do not keep it.
- Google Analytics collects usage statistics on the public site.
- Have I Been Pwned is used to reject passwords that appear in known breaches. Only a short prefix of a hash of your password is sent, never the password.
These providers process data mainly in the United States, and Cloudflare also in other countries where it operates data centres. Those countries may have different personal-data rules from the one you live in; we rely on each provider’s contractual commitments and security measures, and will tell you on request which safeguards apply to a given provider.
5. How long we keep it
- Account, session, cloud project and AI workspace data: for as long as your account exists, or until you delete the project or workspace. Expired sessions and codes are removed.
- Billing records: for as long as needed to manage your subscription and to meet accounting and tax obligations, typically up to seven years.
- AI usage records (model, token counts and estimated cost per call; no message content): for as long as your account exists, for plan limits and cost reconciliation.
- A separate service-wide AI spending ledger retains only a random call identifier, time, accounted amount and budget state, including after account deletion. It contains no account or workspace identifier, email, model, token counts or message content. We retain it to keep already incurred or reserved costs within the service budget and to reconcile costs; deleting an account does not restore spent service funds.
- Diagnostics: currently seven days.
- MCP relay messages: until delivered or expired, typically minutes.
- MCP connection records: 180 days.
- Viewing links: the shared copy until you stop the link or it expires (at most 30 days); the record of the link one week after that.
6. Your choices and rights
You can export every project you own at any time from Studio. Changing or resetting your password ends every other session. To access, correct, stop the use of or delete the personal information we hold, or to delete your account, email us at the address below from your registered address; we will confirm it is you and respond without undue delay, normally within two weeks. Deleting your account removes your cloud projects, AI workspaces and account-linked usage records; the limited service-wide spending ledger described above is retained separately. If you have an active subscription we cancel it when the account is deleted, with no refund for the remaining period. Deleted data can remain in database backups for up to 30 days. Billing records that we are required to keep are retained for the statutory period.
You may opt out of Google Analytics with browser settings or the Google opt-out extension. Blocking the session cookie prevents signing in.
7. Security
Passwords are hashed, sessions carry an absolute expiry, MCP links are tied to your login and can be revoked, and administrative access is limited to the operator. No service can promise perfect security; if we learn of a breach affecting your data we will notify you as the law requires.
8. Children
The Service is intended for professional use and is not directed to children under 16. We do not knowingly collect personal information from children.
9. Changes to this policy
We may update this policy as the Service changes. The date at the top shows the latest revision. For material changes we will post a notice on the site or email registered users before the change takes effect.
10. Contact
Questions and requests about personal information: support@laydyne.com. The Service is operated under the trade name Laydyne.