This policy describes how BingeRead collects, uses, and stores information when you use the Chrome extension and companion web app.
Last updated: September 30, 2026
The Chrome extension uses browser permissions needed to store settings, inject the transcript interface on supported video pages, download user-requested exports, and support account sign-in. Browser alarms retry pending account-library synchronization after temporary network failures or browser restart. We do not use these permissions for unrelated tracking, advertising, or background browsing analysis.
BingeRead may store settings, login state, cached transcript transformations, generated results, History, and reading progress locally in your browser so the extension can work reliably and avoid repeating work. You can remove this local extension data by clearing the extension's site/storage data or uninstalling the extension.
You can view available captions, use Reader, open local History, and use locally cached results without an account. These local actions do not use a Guest trial and do not create a Supabase anonymous user.
Current manuscript generation requires a verified account. The account includes a one-time generation allowance and a weekly allowance for new shared reads; current amounts are listed on Membership. Reading an already unlocked manuscript does not use another weekly unlock.
Where Guest trials are available, BingeRead provides one Manuscript trial and one Summary trial. They are independent, do not refresh, and use the product's default model. A trial is used only when BingeRead successfully delivers a usable server-generated result or eligible shared-cache result for that feature. Loading a result already stored locally does not use a trial. Failed, timed-out, empty, or canceled operations do not use a trial unless a usable result was already committed and delivered.
BingeRead obtains captions from supported videos opened in the extension or selected for processing through the reading service. Following a YouTube channel with automatic processing enabled allows the service to check for new eligible videos and obtain their captions. For requested AI processing, relevant caption text is sent to our backend and model provider. Reading an existing manuscript does not require a new model request.
We do not use transcript text or generated output for advertising. We also do not sell transcript text, generated output, or account data.
An eligible manuscript may be stored on our server and reused for another BingeRead user only in the context of the same source video. Eligible public manuscripts from channels in the service directory can be browsed and searched in Discover. Personal imports and manuscripts marked private are not listed there. The reading library does not publicly attribute a shared manuscript to the account that generated it. Summary output is not part of this shared manuscript library.
You can choose a private manuscript when starting a new reading-service generation. The extension's contribution setting controls future legacy shared-cache contributions. Your private notes and chat are not made public with a manuscript. Choosing not to contribute does not disable shared reading under the current reading-service plan. Turning contribution off does not remove an already shared manuscript solely because you generated it. Shared manuscripts are separated from the contributor's account identity. If an account is later deleted, an otherwise eligible shared manuscript may remain without an author association. Copyright owners, video platforms, and their authorized representatives may request removal by contacting support@bingeread.app.
When multiple eligible manuscripts exist for the same transcript, BingeRead may use the matching full transcript to compare their quality. That transcript is placed in a separate encrypted review payload, is available only to the review worker, and expires within 24 hours. It is deleted after review or expiry and is not available through the shared-cache delivery or admin search interfaces.
Saved private manuscripts and eligible shared manuscripts are stored on the server for account synchronization and the reading library described above. Caption text may also be stored in processing jobs and recovery checkpoints; it is not exposed as a public raw-transcript library. Backend storage includes reading progress, followed channels, language preferences and metadata needed for authentication, trials, usage accounting, billing, abuse prevention, debugging and service reliability. Local browser storage may contain transcripts, generated results, History, settings, login state, and reading progress.
The use of information received from Google APIs will adhere to the Chrome Web Store User Data Policy, including the Limited Use requirements.
Google account information is used only for sign-in, account linking, credit/plan access, security, and support. It is not used for advertising or sold to third parties.
An inactive anonymous Guest account and its associated server-side product data are removed after 90 days without activity. A non-PII anti-repeat marker derived from the abuse-prevention hashes may be retained for up to one year, then deleted. Local extension data remains on the device until it is cleared by the user, removed by browser storage management, or removed when the extension is uninstalled.
A verified account holder may request account deletion after recent reauthentication. We revoke active sessions immediately and provide a seven-day recovery period. After that period, personal profile, identity, and non-required operational data is deleted or anonymized. Shared manuscripts may remain without an author association as described above. Payment, tax, fraud-prevention, chargeback, dispute, and accounting records may be retained where required by law or reasonably necessary to establish or defend legal claims.
You can request account or data deletion, cancel a pending deletion during the recovery period, or submit a privacy question by contacting support@bingeread.app.
We do not sell personal data. We do not use transcript or account data for advertising.
BingeRead is not intended for children under 13. If you believe a child has provided personal information, contact us and we will take appropriate steps to delete it.
We may update this policy as the product changes. If the changes are material, we will update the date above and provide notice through the website, extension, or another reasonable channel.
For privacy requests, contact support@bingeread.app.