Privacy Policy
Last updated: November 16, 2026
Nexus is a read-it-later and knowledge intelligence application. This policy explains what data we collect, how we use it, who else receives it, and how we protect and delete it.
Who we are
Nexus is run by Products Lab XYZ. We are the controller of the personal data described in this policy: we decide why and how it is used. Nexus is run by one person, so your message reaches the person responsible for your data directly. We are not required to appoint a data protection officer.
Contact us at privacy@productslab.xyz. UK and EU representatives: none yet. Before we offer Nexus to people in the UK or the European Economic Area, that is, before we invite or accept anyone who lives there, we will appoint a representative in each, as UK and EU data protection law requires, and name them here. You will then be able to contact a representative instead of us about anything to do with your data.
This policy covers the Nexus website and web app, the Nexus browser extension and the Nexus apps.
What data we collect
- Google account information — your email address, name, and profile picture, received from Google when you sign in.
- What you save — the web addresses you save, the page as the browser extension captured it, the text and metadata we extract, the files you upload, the podcasts you add, and your reading progress and history.
- What you create — notes, highlights, tags, folders, conversations with the AI assistant, and the summaries, knowledge graph entries and insights generated from your library.
- Gmail data — if you connect Gmail: your messages (content, subject lines, senders and recipients), labels and contacts, accessed only with your explicit permission (see Google user data below).
- Calendar events — if you connect Google Calendar: your events (titles, times, places, descriptions and attendees), accessed only with your explicit permission.
- Your AI provider keys — the API keys you save in Settings (see API key security below).
- Sign-in records — each signed-in session records the IP address and the browser user agent it came from. For each browser you sign in from we also keep a one-way hash of a random cookie (
nexus-device), the browser's user agent, and when it was first and last seen, so we can email you when a new browser signs in. - AI usage records — for each AI request made with your keys: the feature, the model, the token counts and the estimated cost, so Settings can show your spending.
- Voice notes — the recording you make, kept on our servers only until it is transcribed, and the resulting text.
- Your interest profile — the topics that come up most in what you save and read, a numeric summary of them, and the scores Nexus gives items for you (see Personalisation and automated decisions).
- Activity records — a log of the actions you take in Nexus (which action, when and on which item, never what you typed) and, for an item you delete, its title, web address and dates.
- Invitations — if we invite you, your email address on our sign-up list, before you create an account.
- Messages you send us — emails to our privacy or legal address, and our replies.
Why we use your data, and our legal bases
- Providing Nexus to you — your account, saving and reading, notes and search, the AI features that run on your keys, Gmail and Calendar if you connect them, the extension and apps, the emails you ask for, showing your AI spending, and ordering your library by what interests you. Legal basis: performing our contract with you (the Terms of Service).
- Keeping your account and Nexus secure — sign-in records, the browser cookie behind new-sign-in emails, rate limits, the activity log, and investigating abuse. Legal basis: our legitimate interest, and yours, in protecting accounts and the service.
- Keeping Nexus working — server logs, error reports, tracing, background-job records, operational alerts and backups. Legal basis: our legitimate interest in finding and fixing faults and recovering from failures.
- The early-access list and the waitlist — storing your email and answers to confirm your place and tell you when it opens: steps you asked us to take before you join. Notes about the features you ticked: your consent, which you can withdraw at any time.
- The home-page experiment — counting which of two home-page headlines leads more visitors to sign up. Legal basis: until the cookie notice described under The home page and the waitlist arrives, our legitimate interest in improving the home page; from then on, your consent.
- Other people's information in your content — our legitimate interest, and yours, in providing the service you asked for (see below).
- The law and legal claims — complying with legal obligations and lawful requests, and our legitimate interest in establishing or defending legal claims.
Where we rely on legitimate interests, we have weighed them against your rights, and you can object at any time (see Your rights). You need a Google account email address to have a Nexus account. Everything else, such as AI keys, Gmail, Calendar and the extension, is optional: without it, only that feature is unavailable.
Google user data
Nexus asks Google only for the access a feature needs, when you first use that feature. Gmail and Google Calendar features are currently available only to the Nexus founder's own account: every other account is asked only for sign-in access.
- Sign-in:
openid,emailandprofile, to identify you and show your name and picture. - Gmail:
https://www.googleapis.com/auth/gmail.modify, requested when you connect Gmail, to show your mail in Nexus, send the emails you write, and carry out the changes you make, such as labels, archiving and read status. - Calendar:
https://www.googleapis.com/auth/calendar, requested when you connect Google Calendar, to show your events, save the events you create or change in Nexus, and receive Google's change notifications so the calendar stays current.
Nexus'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.
- Your Gmail and Calendar data is used only to provide the email and calendar features you use in Nexus. When you use an AI email feature, the messages and events it works on are sent to the AI provider you chose for that mailbox in Email settings (OpenAI or Anthropic), under your own API key, only to produce the result you asked for. Under those providers' standard API terms, they do not use API data to train their models unless you have opted in to that with them.
- We do not sell this data; use it for advertising, including retargeting or interest-based advertising; use it to determine creditworthiness or for lending; or use or transfer it to develop, improve or train generalized or non-personalized AI or machine-learning models.
- We transfer it to no one else, except as needed to provide those features with your consent, for security purposes such as investigating abuse, to comply with the law, or as part of a merger, acquisition or sale of assets after asking for your explicit consent.
- No one at Nexus reads it, except with your permission for specific messages, when needed for security (such as investigating abuse), or to comply with the law.
- You can remove Nexus's access at any time at myaccount.google.com/permissions. Deleting your Nexus account also asks Google to revoke it.
How AI processes your data
Nexus uses AI to provide intelligent features including article triage, summarization, knowledge graph extraction, search, audio, and conversational chat.
- AI features run only on the API keys you save in Settings; Nexus does not use its own provider keys for your requests. Without a key, the AI feature is skipped.
- Each request goes from our servers to the provider whose key it uses, under your account with that provider. A provider receives your data only if you saved a key for it.
- Your content is sent under your own account with that provider, only to produce the result you asked for. How long the provider keeps it is set by your account's terms with that provider.
What each provider receives:
- OpenAI — the default for most features: the text of your articles, PDFs and notes; your chat and search questions with the library passages that answer them; text to read aloud; voice notes, and podcast audio when you have no AssemblyAI key, to transcribe; and the first pages and cover image of a PDF you upload, to find its title.
- Anthropic — the same kinds of text, for any feature where you choose a Claude model, or by default when it is your only key. Not used for search or audio.
- Google Gemini — the same kinds of text, for any feature where you choose Gemini (chat's automatic model choice can pick it), text for search when you choose Gemini for search, and podcast scripts when you choose Gemini voices.
- Perplexity — deep research questions with the library context; chat questions with library passages when it is your only key; and a podcast episode's title, show and a short excerpt, to find similar episodes.
- ElevenLabs — podcast scripts, to voice them, when you choose ElevenLabs voices.
- AssemblyAI — a link to the podcast audio to transcribe: our stored copy, whose link works without signing in, or the publisher's own file.
- Cohere — text for search, including your search questions, when you choose Cohere for search.
Gmail content goes only to the AI provider chosen for that mailbox in Email settings, under your own account with that provider and only to produce the result you asked for: the sender, subject, preview and labels of each message for automatic sorting (with the start of the message when list summaries are on); whole threads for summaries and drafts; your sent mail, to learn your writing style; and calendar events with their attendees, for meeting preparation. When you turn on web search in chat, the provider's own search tool searches the web for your question.
Who is responsible for what. We are responsible for your data while Nexus stores and processes it. When you save a key, you tell Nexus to send each AI feature's input to that provider under your account. The provider then handles it under its agreement with you, not under a contract with us: for most of these providers, their API terms treat them as your service provider and say whether they keep, review or train on what they receive. Please check those terms and your account settings with each provider. In particular, the free (unpaid) tier of Google's Gemini API may use what it receives to improve Google's products and may let people review it; to avoid that, use a Gemini key from a Google Cloud project with billing turned on.
Personalisation and automated decisions
Nexus builds a profile of your interests from what you save and read (the topics that come up most, and a numeric summary of them) and uses it, with AI scores for each item, to order your Triage queue (“Best match”) and to suggest related and trending items. This is profiling, but it only arranges your own library: it is never used for advertising, never shared, and never used to make decisions about you with legal or similarly significant effects. To choose which public news sources appear in Trending, we also combine the most common topics of all accounts, without linking them to anyone. Your profile is deleted with your account.
What you save can reveal sensitive things about you, such as your health, beliefs or politics. Nexus does not try to infer such characteristics about you, and uses your library only for the features you use.
Other people's information in your content
What you keep in Nexus can include information about other people: people named in the pages, files and notes you save (Nexus extracts the names of people, organisations and topics to build your knowledge graph) and, if you connect them, the people you email with and the attendees of your events. We use it only to provide the features you use in your own account, such as showing who you correspond with most, and never to contact those people or to share or sell their information. Because we cannot practically tell each of them, this policy is how we inform them; anyone can exercise their rights by writing to privacy@productslab.xyz.
API key security
The API keys you save for any AI provider are encrypted at rest using AES-256-GCM encryption. Keys are only decrypted at the moment an API call is made and are never logged, displayed in full, or exposed through any interface. The access and refresh tokens Google issues to Nexus are also stored encrypted.
Fetching the pages you save
To read a page or media file you save, our servers fetch it from its own site. Some pages and media need help from another service, which receives only the address or id below, from our servers, never your name, email address or account:
- Jina Reader — the page's web address, when our own reading of the page fails.
- The Internet Archive's Wayback Machine — the page's web address, to find an archived copy.
- Simplecast — a podcast episode's id, to find its audio.
- YouTube — the video's id, for its captions and title, and its web address, to download its audio for transcription.
When you open a saved item, your browser may load its images or media directly from the original site or its content network, which then receives your IP address and browser details, as with any web page. It learns that the request came from Nexus, not which page in Nexus.
Data storage and security
- Nexus runs on Google Cloud in the us-central1 region (United States): the application servers, the PostgreSQL database and Redis. Stored files and backups are kept in Google Cloud Storage. Google Cloud encrypts all of them at rest.
- Your browser, the extension and the apps connect to Nexus over HTTPS (TLS), and our servers use HTTPS to reach the AI providers and the other services named in this policy. A page you save from a plain
http://address is fetched the way that site serves it. - Sign-in uses secure, HTTP-only cookies. Your AI keys and the tokens Google issues to Nexus are encrypted (see API key security), and the keys the extension and the Android app use are stored only as a one-way hash.
- Links to stored files. The audio and images we store for your items, such as an article read aloud, a podcast copy or an article's pictures, are served from long, hard-to-guess links that work without signing in: anyone who has such a link can open that file. Files you upload, such as PDFs, and magazine files are served only through short-lived signed links.
- Only the Nexus operator has access to the servers and the database. We look at your content only when you ask us to (for example, to fix a problem you report), for security, or when the law requires it.
- If a security breach is likely to put your data at high risk, we will tell you without undue delay, and we report breaches to the data protection authorities as the law requires.
International transfers
Nexus is run from Ukraine and its servers are in the United States. If you are in the UK or the European Economic Area, your data is therefore handled outside the UK and the EEA:
- Our service providers — Google Cloud, Sentry and Resend, companies based in the United States, process data for us under data processing terms that include the European Commission's standard contractual clauses and the UK addendum to them; where a provider is certified under the EU–US Data Privacy Framework and its UK extension, that applies too.
- Access from Ukraine — the operator runs Nexus from Ukraine. Neither the UK nor the EU has an adequacy decision for Ukraine; the operator accesses your data only to run and support Nexus, under the protections described in this policy.
- Your AI providers are mostly in the United States. We send them data on your instruction, and your own agreement with each provider, including any data-transfer terms it offers you, governs how it handles that data.
- Jina Reader, the Wayback Machine, Simplecast and YouTube receive only a web address or an id from our servers, never your name, email address or account.
- Telegram, which delivers our operational alerts to the operator, has no data processing agreement with us, so we limit what we send it to an error message and a job reference (see Third-party services for what they can contain).
You can ask for a copy of the safeguards we rely on at privacy@productslab.xyz.
Data retention
- Your library, notes, conversations and the data AI derived from them are kept for as long as your account exists, unless you delete them.
- When you delete an article, all derived data (summaries, knowledge graph entries, embeddings) is permanently removed, and its stored files are queued for deletion.
- Sign-in sessions expire after 7 days without use; sign-in verification codes after 10 minutes.
- Sign-in records (the IP address and browser of each session) are deleted when you sign out of that session, or 30 days after it expires. The list of browsers you have signed in from is kept while your account exists; you can see the devices signed in to your account, and sign them all out, in Settings → Devices.
- The activity log (which action, when and on which item, never what you typed) is kept for 12 months. When you delete an item, the log keeps its title and web address for 90 days, so that a mistaken or disputed deletion can be traced; then they are removed from the log.
- Items you dismiss in Triage are deleted automatically after 30 days. A voice note's recording is deleted from our servers once it has been transcribed; we keep the text. Extension and app keys stop working after 90 days without use, and a removed key is deleted after 30 days. Disconnecting Google Calendar, or removing a mailbox, deletes the events or mail copied into Nexus.
- A home-page visit record is deleted 13 months after the last visit, and a waitlist entry 30 days after you create your Nexus account (see The home page and the waitlist).
- Database backups are kept for 90 days; disk snapshots and database change logs for 7 days. A deleted file, a backup included, stays recoverable for 7 more days, then is gone.
- Server logs are kept for Google Cloud Logging's default period of 30 days. They include, for every request to Nexus, your IP address, your browser's user agent, the address requested and the page that referred you, and also your account's internal id and the addresses of pages you save. Our tracing system, which runs on our own servers, keeps for 60 days a record of each AI request (the feature, model, token counts and cost, with your account's internal id, but not the text) and of background jobs (which can include a job's result or error message). Copies of the tracing records are also in the database backups, so they can last up to 90 days.
- Background jobs (the work Nexus does after you save a page, such as reading and summarising it) keep a record of their input, which can include a saved page's address and content, for 30 days after they finish; a daily clean-up deletes older ones. A job that keeps failing is set aside for the same 30 days.
- Our service providers keep their own records: Sentry keeps error reports, and Resend a record of each email it delivered (address, subject and status), for the periods their services set. Operational alerts stay in the operator's Telegram chat until they are deleted there.
Deleting your account
You can delete your account at any time in Settings (it asks you to confirm with a recent Google sign-in). When you do:
- Removed immediately: your account and everything stored with it, including your articles, notes, highlights, conversations, knowledge graph, insights, the email and calendar data copied into Nexus, your AI keys and usage records, and your sessions and devices. So are your waitlist and early-access entry, your sign-up allowlist entry, pending verification codes, the home-page visit records linked to you, and your background jobs still waiting to run or set aside after failing.
- Your stored files (article images, audio, PDFs and podcast files) are queued for deletion at once, with a second pass six hours later.
- Google access revoked: we ask Google to revoke Nexus's access to your Google account, Gmail and Calendar included, and stop Google Calendar's change notifications to Nexus. If Google cannot be reached at that moment, you can remove the access yourself at myaccount.google.com/permissions.
- Backups: copies in database backups roll off within 90 days; disk snapshots, database change logs and copies of deleted files roll off within 7 days (a deleted backup stays recoverable for those 7 days too). Server logs, tracing records and the records of finished background jobs roll off as described under Data retention.
- Elsewhere: error reports at Sentry and Resend's delivery records roll off on those services' own schedules, and operational alerts stay in the operator's Telegram chat until they are deleted there. Copies on your own devices, such as the extension's sign-in details or an offline copy of your library in a browser, stay until you remove the extension or clear that site's data.
Third-party services
- Google — sign-in, and Gmail and Calendar access when you connect them (see Google user data).
- Your AI providers — OpenAI, Anthropic, Google Gemini, Perplexity, ElevenLabs, AssemblyAI and Cohere, only those you saved a key for (see How AI processes your data).
- Jina Reader, the Wayback Machine, Simplecast and YouTube — fetching pages and media (see Fetching the pages you save).
- Google Cloud Platform — application hosting, file storage, backups and server logs (Cloud Logging).
- Sentry — error tracking and performance monitoring, from our servers and from your browser, the home page included. Reports carry the error, where it happened, the address requested, how long pages took to load, and technical details such as the browser version. Nexus does not attach your name, email address, account id, cookies, request bodies, or AI prompts and answers to them, and background-job ids reach Sentry only as a one-way hash. The text of an error message is sent as it is, so it can include an internal id or the web address of a page involved in the error.
- Resend — sending email: sign-in verification codes, the one confirmation for the waitlist, an alert when a new browser signs in to your account (with the browser's type and IP address), and a notice when a Google account is linked to yours.
- Telegram — operational alerts to the Nexus operator about errors and failed background jobs: the error text (for a failed job, with web addresses removed) and the failed job's id, which can include your account's internal id.
Browser extension permissions
The Nexus browser extension requests access to nexusprod.productslab.xyz to authenticate you and save the pages you choose, and talks to no other server. It reads a page only when you save it: a page save sends its web address, title and canonical address together with the page's HTML as your browser shows it, which can include content you see because you are signed in to that site; saving a link sends only its address. In the browser's local storage it keeps its sign-in key, your email address and name (to show who is signed in), the Nexus address it talks to, and a queue of up to 100 pages waiting to be saved (web address, title and canonical address) while Nexus cannot be reached. Recent save results are kept only for the browser session. The extension's use of the information it receives adheres to the Chrome Web Store User Data Policy, including the Limited Use requirements. It does not collect your browsing history: its access to your tabs is used only to read the page you choose to save and to finish signing in.
Cookies and similar technologies
Nexus uses only its own (first-party) cookies and browser storage. There are no advertising cookies and no third-party analytics; error and performance reports from your browser go to Sentry (see Third-party services). Everything below is needed for something you asked for, except the home-page experiment cookie (see The home page and the waitlist).
- Signing in —
__Secure-better-auth.session_tokenkeeps you signed in and ends 7 days after you last use Nexus;__Secure-better-auth.session_datacaches your session for 60 seconds;nexus-csrf-tokenandnexus-gate-exempt(24 hours each) are part of the sign-in checks. - Account security —
nexus-device, a random id that lets us recognise your browser, so we can email you when a browser we have not seen before signs in to your account. We keep only a one-way hash of it. It lasts one year. - Language —
nexus-locale(enoruk), set only when you choose a language on the home page or the sign-in page. It lasts one year. - App preferences —
nexus-onboarded(which account on this browser finished the welcome steps; one year) and layout choices such as the reader font or collapsed panels, kept in your browser's local storage until you clear it. - Offline reading — if you turn on offline mode, the text of your library is stored in this browser so you can read without a connection. Signing out does not remove it; it is replaced if a different account signs in on this browser. To remove it, clear this site's data in your browser settings.
- The home-page experiment —
nexus-visitorand a note in local storage; see The home page and the waitlist.
Two more cookies are set only on the founder's own account, for its extra checks.
The home page and the waitlist
- The headline experiment. We are testing two versions of the home page headline. When you first visit the home page signed out, we set
nexus-visitor, a signed first-party cookie holding a random visitor id and the headline you were shown, so you keep seeing the same version, and the page keeps a note in your browser's local storage (nexus-landing-view) so a visit is counted at most once a day. The cookie lasts about 400 days and is never shared. It is not needed to show you the page: it is there to count which headline works. Today it is set on your first visit, before you are asked. We are adding a cookie notice: once it is live, this cookie is set only if you accept it, and if you decline, nothing is stored or counted. - What we record about a visit. The visitor id, the headline shown, when you first and last visited, the site that sent you (its host name only, never the full address), a campaign tag if the link had one and, if you answered them on the sign-up form, your answer to “Which sounds more like you?” and the features you ticked. The visit record holds no IP address or browser details. If you then sign up, we link this record and the headline you saw to your account, to count which headline leads to sign-ups. We delete visit records 13 months after the last visit.
- Server logs. Like every request to Nexus, a visit also appears in our server logs, with your IP address, browser user agent and the full address of the page that referred you, kept for about 30 days (see Data retention).
- The early-access list and the waitlist. If you leave your email, we store it with the headline you saw, your optional answer to “Which sounds more like you?” and the features you asked to hear about. We use it only to write to you about Nexus: one confirmation now, a note when your place opens, and a note when a feature you ticked is ready. Every note about a feature will have an unsubscribe link.
- Retention and deletion. A waitlist entry is kept until 30 days after you create your Nexus account, or until you ask us to remove it. Email privacy@productslab.xyz and we will delete your waitlist entry and any visit record linked to it; deleting your Nexus account deletes them too.
Children
Nexus is not for children. You must be 18 or older to use it, and we do not knowingly collect data from anyone under 18. If you believe a child has given us data, email privacy@productslab.xyz and we will delete it.
Your rights
Depending on where you live, you have the right to:
- access your data and get a copy of it;
- have data corrected if it is wrong or incomplete;
- have your data deleted (you can delete your whole account in Settings);
- restrict how we use your data while a concern is looked into;
- receive the data you gave us in a machine-readable format, or have it sent to another service. Settings → Import & export downloads your saved articles with their addresses, titles, details and text; for a full copy, including notes, highlights and conversations, email us;
- object to our use of your data based on legitimate interests, and at any time to marketing;
- withdraw your consent at any time where we rely on it (the notes about features you ticked and, once the cookie notice is live, the home-page experiment cookie), without affecting what we did before.
To use any of these rights, email privacy@productslab.xyz from the address on your Nexus account. If you write from another address, we may ask you to confirm the request from your account address or by signing in, so that we never give your data to someone else. It is free. We reply within one month; if a request is complex we may need up to two more months, and we will tell you within the first month if so.
Complaints. If you are unhappy with how we handle your data, please tell us first at privacy@productslab.xyz: we will acknowledge your complaint within 30 days, look into it, and tell you the outcome without undue delay. You can also complain to a data protection authority: in the UK, the Information Commissioner's Office (ico.org.uk/make-a-complaint); in the European Economic Area, the authority where you live or work (list of authorities); in Ukraine, the Ukrainian Parliament Commissioner for Human Rights (ombudsman.gov.ua).
Changes to this policy
When we change this policy we update the date at the top. If a change is material, for example a new kind of data, a new purpose or a new service provider, we will tell you by email or in Nexus at least 14 days before it takes effect, as the Terms of Service say. We will not start using data we already hold, including data from Google, in a new way without telling you first and, where the law or Google's policy requires it, asking for your consent.
Contact
For privacy-related questions or data deletion requests, contact us at privacy@productslab.xyz.