Legal
Privacy notice
Effective 4 October 2026 · version 2.1
This notice explains what scrubbr.ai collects when you sign in with Google or Microsoft and connect Gmail, Google Calendar, Google Chat, Slack, Outlook, or Teams to your scrubbr.ai account; how we use that data; and the rights you have over it.
1. Who we are
Scrubbr.ai is operated by Scrubbr.ai LLC, a California limited liability company. We are the controller of personal information you provide to use scrubbr.ai. You can reach us at privacy@scrubbr.ai.
2. Information we collect
Account information
Name, email, profile photo, organization, and any preferences you set. We receive these from Google or Microsoft, depending on which account you sign in with.
Gmail content (with your permission)
Once you authorize Gmail through Google OAuth, scrubbr.ai reads message metadata (sender, recipient, subject, timestamps, labels) and message body content so the product can prioritize, summarize, and draft replies. Attachment content is never stored in our database: when you open an attachment, we fetch it from Gmail and hand it straight to your browser without keeping a copy, and attachments you add when composing are streamed straight to Gmail without being persisted. The one exception is small images embedded inline in an email's body, which are stored as part of that message.
Slack content (with your permission)
Once you authorize Slack, scrubbr.ai reads message metadata and message body content from channels and direct messages you have access to. Scrubbr.ai uses your personal user token, not a bot token; messages you send through scrubbr.ai appear in Slack under your name. Attachment content is never stored in our database — links to Slack-hosted files are preserved so you can open them in Slack, but the file bytes are not copied.
Google Calendar content (with your permission)
Once you authorize Google Calendar, scrubbr.ai reads event metadata (titles, attendees, times, locations, descriptions) so the product can surface upcoming meetings, generate prep context, and reflect schedule conflicts in your feed. Files linked to events stay in their original location (Drive, web links) and the file content is not stored in our database.
Google Chat content (with your permission)
Once you authorize Google Chat, scrubbr.ai lists the spaces and direct messages you belong to, reads message text in those spaces, and sends replies on your behalf when you approve a draft. Attachment content in chat messages is never stored in our database — file references stay in Google Chat where the original was posted. If you message someone you have never chatted with before, scrubbr.ai creates a new 1:1 direct message space on your behalf.
Derived content
Summaries, draft replies, priority tiers, and embeddings (numeric vectors used for retrieval) generated by AI from the content above. Derived content is associated with your account and used only to power your scrubbr.ai experience.
Structured facts (the "knowledge layer")
Scrubbr.ai also extracts short, structured facts from the content above so the assistant can recall them later — for example, "[name] is the CFO at [company]", "[name] owes [name] the Q3 numbers by Friday", "the kickoff meeting is at 3pm Tuesday", or a link to a document someone shared. These facts are derived from the same workspace content covered above, scoped to your account, used only to power your scrubbr.ai experience, and not used to train AI models. Each fact carries a confidence score and is automatically superseded or marked resolved when newer information makes it stale. They are stored in the same database as your other content and protected by the same row-level security and access controls.
Operational data
Logs, error reports, and product usage events. We do not include message content in logs.
Communications with us
Messages you send to support, sales, or any of our team email addresses.
2b. Google OAuth permissions
When you connect your Google account, scrubbr.ai requests the following permissions. We request only what is necessary to deliver the features you use.
| Permission | Why scrubbr.ai needs it |
|---|---|
| gmail.modify | Read, classify, and surface messages; archive and mark read on your behalf; store AI drafts in your Gmail drafts folder; send approved replies as you. We do not permanently delete messages and we do not send without an explicit action from you. |
| gmail.settings.basic | Read your Gmail send identities (including the signatures you have configured for each), and your vacation responder status. Used to append your real Gmail signature to messages scrubbr.ai sends, to reply from the correct alias when a thread arrived at one of your alternate addresses (e.g. founder@), and to surface a "you are set to out of office until [date]" reminder on your dashboard. Scrubbr.ai uses this permission read-only and does not modify your Gmail settings. |
| calendar.events | Read upcoming events for schedule view and pre-meeting briefings; create, update, and delete events when you schedule or cancel from inside scrubbr.ai. |
| calendar.calendarlist.readonly | Read the list of calendars you have access to in Google Calendar — your primary calendar, team calendars, shared client calendars, holiday calendars, and any conference-room or resource calendars. Used to sync events from each calendar into scrubbr.ai's unified view and to power the "calendars" toggle list in settings, where you can mute noisy calendars. We do not modify calendar memberships. |
| calendar.settings.readonly | Read your Google Calendar preferences — specifically your working hours and default timezone. Used so that when scrubbr.ai proposes meeting slots ("find me time this week"), the proposals respect your actual working hours and timezone instead of guessing. Read-only — scrubbr.ai never modifies your calendar settings. |
| admin.directory.resource.calendar.readonly | Read the conference rooms (resources) configured in your Google Workspace organization, so we can suggest them when you pick a meeting location. We access only room metadata (name, email, capacity, features, building, floor) — no user data, no calendar event content. Results are rendered live in your browser and not stored on our servers. |
| chat.spaces.readonly | List the Google Chat spaces and DMs you belong to so scrubbr.ai knows which conversations to sync. No message content is accessed through this permission. |
| chat.messages | Read chat messages for the unified inbox; send approved draft replies as you. Message text only — attachment content is not stored in our database. |
| chat.spaces.create | Create a new 1:1 direct message space when you send a chat message to someone you have never chatted with before. Only triggered by an explicit send action from you. |
| chat.memberships | Read the membership (list of people) in chat spaces you already belong to, and add, remove, or update members of those spaces when you take an explicit action in scrubbr.ai (e.g. "add Jane to this space" from the roster panel, or "leave this space"). Used to show the member roster + power @mention autocomplete, and to mirror Google Chat's native add/remove-member sidebar inside scrubbr.ai. We do not list memberships of spaces you are not in, we never modify membership without an explicit action from you, and we do not persist the roster on our servers — results are rendered live in your browser only. |
| chat.messages.reactions | Read and add emoji reactions on Google Chat messages — you can tap an emoji to react, and reactions other people leave on chat messages are visible to you. Reactions are scoped to messages in spaces you already belong to. We never modify the content of a chat message; only your reactions. |
| directory.readonly | Read the names, email addresses, and basic organizational info (job title, department, photo) of users in your Google Workspace organization. Used only to suggest your teammates when you click "view teammate calendar" so you can check their availability while scheduling. We do not access phone numbers, addresses, login activity, or any other extended profile data. Results are rendered live in your browser and not stored on our servers. |
| contacts.other.readonly | Read your "other contacts" — people you have emailed but have not added to your saved Google contacts. Used only to display profile photos for the senders in your unified inbox, so an avatar appears next to each message instead of an initials placeholder. We read profile photo URLs only. No image bytes ever touch scrubbr.ai's servers — your browser fetches the photo directly from Google's CDN. We do not access phone numbers, addresses, or any other contact details. |
You can review and revoke these permissions at any time via your Google account permissions page or via scrubbr.ai settings → integrations.
2c. Google API Services User Data Policy
Scrubbr.ai's use and transfer to any other app of information received from Google APIs will adhere to the Google API Services User Data Policy, including the limited use requirements. Specifically: we use Google data only to provide user-facing features in scrubbr.ai; we do not use it for advertising; we do not sell it; we do not allow humans to read it, except where you have given us your affirmative agreement to view specific messages, where it is necessary for security purposes, or where required to comply with applicable law.
3. How we use your information
- To operate scrubbr.ai: prioritization, drafting, calendar prep, search, and your settings.
- To improve the specific features we provide to you, based on your usage of those features. We do not train AI models on your content (see section 6).
- To communicate with you: service notices and your support requests. We do not send marketing without consent.
- To keep the service safe: abuse detection, integrity, and legal compliance.
5. How long we keep it
Message and calendar content is kept for the life of your account by default, so search and drafting can use your history. You, or your company administrator, can set a shorter window in Settings, from 30 days to 10 years; a nightly job then deletes messages older than the window, together with their search embeddings, extracted commitments and past calendar events. Some derived records, such as extracted facts, meeting-note summaries and voice assistant logs, are not yet covered by the window; they are deleted when you delete your account. Audit log entries (sign-ins, integration changes, sends; they include IP address and browser, never message content) are kept for 365 days as security evidence.
If you cancel or delete your account, we delete your account data within 30 days, except where retention is required by law. You can request export or deletion at any time at privacy@scrubbr.ai.
6. AI training and your data
Scrubbr.ai does not use your content to train AI models. Every model request goes through our AI gateway with a no-training, no-data-collection policy, so it is routed only to providers whose data policies prohibit training on it. Voice input is transcribed by your browser; in Chrome, the browser's speech service (operated by Google) processes the audio under the browser's own terms, and scrubbr.ai receives only the text.
7. How we protect it
Data is encrypted in transit (TLS 1.2 or higher) and at rest (AES-256 with provider-managed keys). Production access is limited to our two founders. Our security policies are aligned with NIST SP 800-171 Rev. 2. Scrubbr.ai has completed Google's CASA Tier 2 security assessment. Scrubbr.ai is not yet SOC 2-attested; a SOC 2 Type I audit is engaged with TAC Security, and the letter of engagement is available on request.
We have not built any internal tool for reading customer messages. The database itself can be reached by our two founders, for incident response and approved repairs; that limit is enforced by policy, not yet by encryption, and reads made directly in the database console are not individually logged today. Multi-factor authentication protects our AWS root and administrator accounts, and enforcing it on every human account with production access is in progress. We access the contents of your messages only with your explicit approval, except where we must act on a security incident or a legal obligation. For data obtained through Google APIs, human access is further limited by Google's API Services User Data Policy.
8. Where your data lives
Scrubbr.ai stores data in the United States. AI processing is performed in the United States by Anthropic, OpenAI, and Google. If you are located outside the United States, your data will be transferred to and stored in the United States under appropriate safeguards.
9. Your privacy rights
Depending on where you live, you may have rights to access, correct, export, or delete your information, and to object to or restrict certain processing. Contact privacy@scrubbr.ai to exercise any of these.
We do not sell your personal information. We do not share it for cross-context behavioral advertising.
10. Cookies and tracking
Scrubbr.ai uses a small number of strictly necessary cookies to keep you signed in and to remember preferences. On this marketing site (scrubbr.ai) only, PostHog sets first-party analytics cookies and browser storage so we can count visits and see which pages and buttons work; the app does not use them. We do not use third-party advertising trackers on the marketing site.
11. Children
Scrubbr.ai is not directed to children under 16. We do not knowingly collect data from anyone under 16.
12. Changes to this notice
When we make material changes, we will update the effective date and notify active customers by email. Previous versions are available on request.
13. Contact
Questions, requests, or complaints: privacy@scrubbr.ai. For security reports: security@scrubbr.ai. For general support: support@scrubbr.ai.