1. Introduction
PatentNode is a research and demonstration product for US patent practitioners. This Privacy Policy explains what we collect when you visit the site, sign in, request early access, or run an Office Action. Most visitors use a sample or demo experience in the browser; invited early-access users may submit a real, public Office Action, which is processed by the third-party AI providers listed below to produce the analysis. We never ask for, and you must not submit, privileged or confidential matter.
2. Information we collect
When you sign in with Google: your name, email address, and profile image, as provided by Google OAuth, stored to maintain your account. When you request early access: a record that you requested access, plus any details you choose to provide (your role, firm, USPTO registration number, whether you have a public Office Action to test, and any note), linked to your account. We do not collect a file at this step. When you run an Office Action (early access): the Office Action text and the related public application data you submit are processed to produce your analysis (see processors below). Submit public matters only. When you use the site: standard session cookies (set by NextAuth) and basic request logs (IP, user agent, timestamps) retained for operational and security purposes.
3. How we use information
We use your account information to keep you signed in, show you which review threads you have access to, and contact you about service updates. We use early-access requests to contact you when access opens. We do not sell or rent your information to third parties.
4. Third-party processors
Service providers (sub-processors) we use: • Google: sign-in identity • Vercel: web hosting and edge delivery • Cloudflare: traffic routing and protection (requests may transit Cloudflare) • Neon: managed Postgres database (account and run metadata) • Resend: transactional email (for example, access notifications) When you run an Office Action (early access), its text is additionally processed by: • OpenAI and Anthropic: model inference on the submitted text • Modal: serving of our in-house models Where offered, these are used under Zero Data Retention (ZDR) terms, meaning prompts and outputs are not retained for model training or beyond the operational window needed to return the response. If and when paid plans open, Stripe processes payment information; we never receive your full card details. Note on in-house models: PatentNode's specialized in-house models were trained only on public USPTO records and synthetic examples; no customer-uploaded content or firm work product was used.
5. Cookies
We use first-party cookies strictly necessary for the site to function; most importantly a session cookie set after sign-in. We do not currently use third-party advertising or analytics cookies.
6. Data retention
Account information is retained while your account is active. You may request deletion at any time using the contact address below; on request we delete your account and associated review marks within a reasonable period, subject to any legal retention obligations.
7. Your rights
Depending on your jurisdiction, you may have the right to access, correct, port, or delete personal information we hold about you, and to object to certain processing. To exercise these rights, contact us at the address below.
8. Contact
Questions or requests regarding this Privacy Policy: privacy@patentnode.com. We will update this Policy when our practices change. The "Effective" date at the top reflects the most recent update.