Privacy at Sixen
Last updated: 1 October 2026
Sixen helps you understand the people you work with. It reads the conversations you choose to bring in, builds a short, correctable read of each person, and suggests how to approach them. This page explains what Sixen reads, what it keeps, where, for how long, and how you delete it.
Who we are. Sixen is run by Elad Hackim, who is responsible for your data (the "controller"). Contact: support@sixen.io.
Sixen is invite-only. Right now only people we invite can use it.
1. What Sixen reads, and only when you ask
Your Google account (sign-in). When you sign in with Google, we receive your name, email address and profile picture, used to identify your account.
Google Drive: only the files you pick. Sixen asks Google for access to only the files you choose in Google's file picker. It cannot see the rest of your Drive. We read the text of each transcript you pick (Google Docs, Word, plain text or VTT captions). We do not keep a Drive access token on our servers: the connection lives in your browser, and "Disconnect Drive" ends it.
Chat exports you upload (for example, WhatsApp). When you upload a chat export, your browser opens the file on your device and sends only the text of the most recent 3,000 messages. Photos, voice notes, videos and other attachments are never uploaded. Your browser also counts, across the whole file, how many messages each sender wrote each month, who starts conversations, how many messages the file holds, and the date of its first message. For the older part of the file, only those counts, the senders' names they belong to, and those dates are sent, never its text. Sixen uses the names only to match the counts to you and to the people you confirmed, and then keeps the counts without the names: yours, everyone else's together, and, in a group chat, those of each person you confirmed. Until you confirm who spoke, these counts wait in that browser tab, like the text.
Transcript files you upload from your device. You can also upload a transcript file (VTT captions, Word or plain text) exported from your meeting tool. Your browser reads the file on your device and refuses audio and video files before reading them. Only the text is sent, and it is handled exactly like a transcript picked from Drive.
If Sixen needs you to confirm who spoke before it can analyze an uploaded chat or transcript, the text waits only in that browser tab until you do.
No audio, ever. Sixen works only with text. It never uploads, stores or analyzes audio, voice or video, and it never tries to infer emotions.
2. What Sixen keeps
- Your account: name, email, settings, the work style you set for yourself, and when you accepted these terms.
- People: the label you give each person (for example initials or a role), their read, and your edits to it.
- Moments: for each observation, a short quote (at most 300 characters) from the conversation it came from, the observation itself, and the date. Every statement about a person is tied to these moments, so you can check it.
- Conversations: each imported conversation's title, date, type, and the people you confirmed in it. For a chat, also the counts described in section 1 (messages per month, who starts conversations, the total number of messages and the first date: numbers and dates only, never text or names; the per-month and who-starts counts are kept only after you confirm who spoke), and the follow-ups and open questions Sixen noted, written in Sixen's own words (at most 280 characters each, never a quote). Sixen does not keep the full text of your transcripts or chats. The text is held only while it is being analyzed and is deleted as soon as that finishes (or fails).
- Daily prep: your prep notes.
- Push notifications, if you turn them on: your device's notification token.
- Usage records: counts and costs of processing per account, used to enforce usage limits. They never contain the content of your conversations.
Sixen uses no advertising and no analytics or tracking scripts. The only third-party code loaded in the app is Google's sign-in and file-picker code.
3. How your conversations are analyzed
Sixen uses Google's Vertex AI, in the EU, to analyze the conversation text you import. Before analysis, each speaker's name is replaced with the label you chose for them, and only people you have confirmed get a read. (Names that people mention inside the conversation itself may remain in the text.) Google does not use this text to train its models.
Google runs automated abuse checks on the text it processes. If a check flags a request, Google may keep that request for up to 90 days, in the EU, under Google's own policy. We can't see or delete those copies.
4. Where your data lives
Sixen runs on Google Cloud (Firebase) in the EU. Our database and our servers are in Belgium (europe-west1), and text analysis runs in Google's EU region. Our operational logs record only identifiers, counts and error codes, never conversation text.
5. What Sixen never does
- It never sells your data or shares it for advertising.
- It never uses your data, or data from Google, to train AI models.
- It is never used to make decisions about anyone's employment: no hiring, firing, pay, promotion or performance ratings, no rankings of people, and no export to an employer.
- Nobody at Sixen reads your conversations or people's reads, except when you ask us to (for example, to help with a problem) or when the law requires it.
6. How long Sixen keeps your data, and how you delete it
Retention period. You choose 30, 60 or 90 days (90 by default) in Settings. Once something reaches the end of its period, Sixen stops showing it and deletes it automatically, usually within a day.
| What | How long it is kept |
|---|---|
| Conversation text | Only while it is being analyzed. Any text left over is deleted automatically after 3 days |
| Moments from a meeting | Your retention period, counted from the meeting date |
| Moments from a chat, and the chat's record | Your retention period, counted from the day you uploaded it |
| A chat's activity counts (messages per month, who starts conversations, total messages, first date) | Same as the chat's record: your retention period, counted from the day you uploaded it |
| Follow-ups and open questions from a conversation | Same as that conversation's record. They are replaced each time the conversation is read again |
| A meeting's record (title, date, people) | Your retention period, counted from the meeting date |
| A conversation waiting for you to confirm who spoke | 16 days |
| A conversation you removed or that was not imported | 14 days |
| Daily prep | 14 days |
| Monthly usage totals (number of analyses and their cost, never content) | Until you delete everything |
| Daily and weekly counters that enforce usage limits (counts only) | Up to 15 days. They expire on schedule even after "Delete everything", so limits can't be reset by deleting |
| People and their reads | Until you delete the person or your account. Automatic expiry for people is being added; this line will change when it is |
| Per-request cost records (your account ID and a cost figure, never content) | Kept for cost accounting. They are not yet deleted automatically, and "Delete everything" does not remove them, because they hold no content |
Delete it yourself, at any time:
- Delete a conversation: removes it and the moments that came from it.
- Delete a person: removes their read, and you can choose to delete all moments and prep about them too.
- Delete everything: Settings → Delete all data removes all your account data at once: people, conversations, moments, prep, devices and your monthly usage totals. You can also delete your sign-in account at the same time.
- Disconnect Drive at any time in Settings, or remove Sixen's access from your Google Account at https://myaccount.google.com/permissions.
7. The people in your conversations
Your conversations include other people. Sixen reads them only to help you understand and work with them. Reads are tentative and meant to be corrected, never final judgments. If you bring other people's conversations into Sixen, make sure you are allowed to, and tell them where your workplace or local law requires it. Anyone who believes Sixen holds information about them can contact support@sixen.io.
8. Google user data
Sixen's use and transfer of information received from Google APIs to any other app will adhere to the Google API Services User Data Policy, including the Limited Use requirements. Data from Google is used only to provide the features described on this page. It is not transferred to anyone else except as needed to run those features (our cloud provider, Google), to comply with the law, or for security. It is never used for advertising, and never used to train AI models.
9. Your rights
You can access, correct, export or delete your data. You can view, correct and delete it yourself in Sixen. For a copy of your data, or anything else, write to support@sixen.io. If you are in the EU or UK, you can also complain to your data protection authority.
10. Changes
If we change this policy, we will update the date above and tell you in the app before the change takes effect.