Privacy & Terms

This site is a public-interest research tool hosting U.S. government documents released under FOIA. We collect as little data as possible and store no personally identifiable information in our application databases.

What we collect

Search queries

When you search the database, the search term and a timestamp are recorded. No IP address, browser fingerprint, or other identifying information is stored. Aggregate search terms are used to display "Top searches" on the homepage. We may review search patterns in aggregate to improve the site and identify research trends.

Page views

When you view a report page, the page path and a timestamp are recorded. No IP address or identifying information is stored.

Aggregate engagement events

On a few pages (currently the Wolff–Epstein email archive and the deposition transcript viewer), the site records anonymous, binned engagement events so we can tell which features readers actually use — for example, whether an audio clip was started, how far down the page someone scrolled (bucketed at 25%/50%/75%/100%), and roughly how long they spent on the page (bucketed at 5s/15s/30s/1m/2m/5m/10m+). No IP address, browser fingerprint, or per-session identifier is stored. A short visitor hash is computed from a daily-rotating server-side salt; because the salt is destroyed and replaced every 24 hours, no visitor can be correlated across days. We use this only to understand aggregate usage patterns of the site itself, never to profile individuals.

AI research assistant

Questions submitted to the AI research assistant are processed in real-time and cached briefly in memory to speed up repeated queries. No questions are permanently stored in any database, and no IP addresses or identifying information are recorded. If you choose to share a conversation, the messages are stored with a random ID. Shared conversations are kept permanently so links do not break. Do not enter personal information — queries are sent to third-party AI services (see below).

Chat history

Your recent chat conversations are saved in your browser's localStorage so you can revisit past chats. This history data is stored entirely on your device and is not sent to our servers. (Your questions are of course sent to our server for AI processing when you ask them — but the saved history sidebar is purely local.) You can delete individual chats or clear all history at any time. If you use private/incognito mode, no history is saved. If you choose to share a conversation, a copy is stored on our server with a random link ID.

Comments

If you leave a comment on a report page, your chosen display name and comment text are stored. No IP address is stored. Rate limiting is handled using a temporary in-memory hash that is discarded within 30 minutes and cannot be reversed to identify you.

Server logs

Like virtually all web servers, our infrastructure (nginx) produces standard access logs that contain IP addresses, timestamps, and requested URLs. These logs are used for security monitoring and are subject to standard log rotation. We do not routinely review individual entries in these logs except to investigate security incidents or abuse. These logs are managed at the infrastructure level and are not part of our application databases.

What we do NOT do

• We do not set any cookies. Our application sets zero cookies. Cloudflare (our CDN) may occasionally set a security cookie (__cf_bm) for bot detection; we have no control over or access to this.
• We do not use analytics services (no Google Analytics, no tracking pixels).
• We do not sell, share, or disclose any data to third parties.
• We do not require accounts, logins, or email addresses.
• We do not build user profiles or track individuals across visits.

Third-party services

• Cloudflare provides CDN and DDoS protection. Cloudflare may set a security cookie and processes requests through its network. See Cloudflare's privacy policy.
• OpenRouter processes AI research assistant queries. Questions are sent to an AI model via OpenRouter's API. See OpenRouter's privacy policy.
• API access by AI agents. This site provides a public REST API for document search and retrieval. Third-party AI assistants (such as ChatGPT Custom GPTs, Claude, and others) may access this API on behalf of their users. API requests are subject to rate limiting and standard server logging as described above. No API keys or user accounts are required.

Your rights

Under the EU General Data Protection Regulation (GDPR) and California Consumer Privacy Act (CCPA/CalOPPA), you have the right to know what data is collected about you. Since we do not store IP addresses or any personally identifiable information in our application databases, there is no personal data to access, correct, or delete. If you have concerns about data in server infrastructure logs, contact us via the contact form.

Content disclaimer

All documents hosted on this site are U.S. government records released under the Epstein Files Transparency Act (Public Law 119-38, signed November 19, 2025). The original documents are in the public domain. Our analysis and commentary are provided under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 (CC BY-NC-SA 4.0) — you may freely reuse and adapt our work, provided you credit the source, do not use it for commercial purposes, and share any adaptations under the same terms. This site is an independent research project and is not affiliated with any government agency, law firm, or news organization.

Terms of Use

By using this site you agree to the following:

• No personal information. Do not enter personal information (yours or anyone else's) into the search or AI assistant. These features are for researching publicly released government documents, not for personal queries.
• AI-generated content. All analytical text and AI assistant responses on this site are generated by artificial intelligence and may contain errors, hallucinations, or misinterpretations. Always verify claims against the linked EFTA source documents before citing or relying on any finding.
• Public domain documents. The underlying EFTA documents are U.S. government records in the public domain. Our analysis and commentary are provided under CC BY-NC-SA 4.0.
• No warranty. This site is provided "as is" for public-interest research purposes. We make no guarantees about accuracy, availability, or completeness. We are not liable for any decisions made based on information found here.
• Victim privacy. If you encounter information that may identify a victim of trafficking or abuse, do not share or republish it. Report it via the privacy/takedown form so we can review and redact if necessary.
• No commercial use. This site and its analysis may not be used for commercial purposes without permission, consistent with our CC BY-NC-SA 4.0 license.

Changes

This policy may be updated. The current version was published on May 6, 2026.