Parallax (\u201Cthe Extension\u201D) is a Chrome browser extension that detects rhetorical framing, logical fallacies, selective evidence, and persuasion techniques in articles. This policy describes what data we collect, why, and how we protect it.
1. Data We Collect
a) Personally Identifiable Information
Email address \u2014 collected when you sign in via magic link. Used for authentication, account identification, and (for Pro and Max subscribers) sending the weekly Info Diet email digest.
Email preferences \u2014 alongside your email we store a flag indicating whether you receive the weekly Info Diet digest. New Pro and Max accounts default to opted-in; you can unsubscribe with one click from any digest email.
b) Authentication Information
JWT token \u2014 stored locally in chrome.storage.local to maintain your logged-in session. We never store passwords.
c) Website Content
Article text \u2014 when you initiate a scan, the visible article text on the current page is sent to our server for AI analysis. We do not collect text without your explicit action.
d) Anonymous Identifiers
Anonymous user ID \u2014 a randomly generated UUID assigned to unauthenticated users for community voting and quota tracking. This ID is not linked to any personal information.
e) Web History
Article URLs — when you scan a page, the URL is sent to our server for analysis, engagement tracking, community voting, and share link generation. We only collect URLs of pages you explicitly scan, not your general browsing history.
f) User Activity
Engagement events — we record which annotations you interact with (e.g. clicking for more detail, voting on an annotation, or dismissing it). These events are tied to your anonymous user ID, not your email or identity.
Article-level framing scores — the overall framing score and technique breakdown for articles you scan are stored to help calibrate our scoring algorithm across news sources.
g) AI Chat Transcripts
Chat transcripts — when you use the AI chat feature (available to Pro and Max subscribers), the questions you ask and the AI’s responses are stored on our servers without any personal identifiers. We use this data to improve answer quality, identify failure modes, and detect abuse. Free and unauthenticated users do not have chat access, so no chat transcripts are stored for them. To request deletion of your chat history, email privacy@parallaxreader.com.
2. Data We Do NOT Collect
Health information
Financial or payment information (Stripe handles all payment processing directly; we never see or store card details)
Location data
General browsing history (we only store URLs of pages you explicitly scan)
Time spent on pages, mouse movements, scrolling, or keystrokes
3. How We Use Your Data
Email & auth tokens \u2014 to authenticate you and manage your account (free or Pro plan).
Article text \u2014 sent to our analysis API to detect rhetorical techniques. Text is processed in real time and is not permanently stored on our servers.
Anonymous ID & votes \u2014 to power community voting on detected annotations without identifying you personally.
Scan quota \u2014 to enforce usage limits (3 free scans/month, 30/month for Pro, or 500/month for Max under our fair use policy).
Engagement analytics \u2014 to understand which detections are useful and improve our analysis accuracy over time. All analytics are anonymised.
Domain-level scoring data \u2014 to calibrate framing scores relative to each news source, making scores more meaningful and comparable.
4. Data Storage
All user data (auth token, email, quota, analysis results, votes) is stored locally on your device using chrome.storage.local. Our backend server processes article text for analysis but does not retain article content after the response is delivered. Anonymised engagement analytics and article-level scores are stored on our server to improve the service.
Railway \u2014 hosts our backend API. No user data is permanently stored on Railway infrastructure beyond transient request processing.
We do not sell, transfer, or share user data with any other third parties.
6. Data Transfer & Sale
We do not sell or transfer user data to third parties, apart from the approved use cases described above.
We do not use or transfer user data for purposes unrelated to the Extension\u2019s single purpose of epistemic annotation.
We do not use or transfer user data to determine creditworthiness or for lending purposes.
7. Remote Code
Parallax does not load or execute any remotely hosted code. All JavaScript is bundled within the extension package. Our server returns only JSON data (analysis results), never executable code.
8. Data Retention & Deletion
Local data is retained until you uninstall the extension or clear it manually. Server-side account data (email, plan status, chat transcripts) can be deleted upon request. To delete your chat history, your account, or any other data we hold about you, email privacy@parallaxreader.com. Weekly digest emails can be stopped at any time using the one-click unsubscribe link in any digest email — your preference is honoured immediately.
9. Fair Use Policy
All paid plans are subject to reasonable usage limits to ensure service quality for all users. The Max plan includes up to 500 scans per 30-day rolling period. The Pro plan includes up to 30 scans per 30-day period. If you reach your limit, scanning will resume automatically when your usage window resets. These limits may be adjusted from time to time.
10. Your Rights
You may request access to, correction of, or deletion of your personal data at any time by contacting us at the address below.
11. Changes to This Policy
We may update this policy from time to time. Material changes will be communicated through the Extension or on this page. The effective date at the top reflects the most recent revision.