Transparency

Victim & Minor Protection

This project maintains a strict prohibition against identifying victims or minors. Our AI analysis prompts explicitly forbid un-redacting names, reversing government redactions, or attempting to identify individuals protected by law. This applies to all automated processing including GPT-4o summaries and entity extraction.

PII Protection

Personally Identifiable Information (PII) of non-public figures is handled with care. When our systems detect sensitive personal data (phone numbers, addresses, financial details) belonging to private individuals, that content is flagged for review. We rely on government redactions as the primary protection layer and supplement with AI-assisted screening.

AI-Assisted Review

Every document in the archive passes through an AI enrichment stage (OpenAI GPT-4o) that generates summaries and extracts entities. These AI prompts are configured to respect existing redactions and avoid speculative identification. AI summaries are navigation aids and may contain inaccuracies — users should always verify against the original source document.

Manual Review Process

Content flagged by users or automated systems undergoes manual review. We prioritize reports involving exposed victim identities, incorrect redactions, or sensitive personal information. Removal requests are typically acknowledged within 72 hours.

Official Government Sources

Every document in this archive originates from one of three official government sources. We do not host leaked, stolen, or unofficial materials. Each document links back to its original source URL for independent verification.

6-Stage Processing Pipeline

Documents pass through a structured ingestion pipeline to make them searchable and analyzable. No content is altered — original files are preserved alongside our processed versions.

Content Removal Process

Epstein Suite indexes publicly released government documents. If you are a rights holder and believe that content hosted on this site infringes your copyright or should be removed for legal reasons, we have a structured process for handling takedown requests.

How to Submit a Takedown Request

1. Identify the content. Provide the URL(s) of the specific page(s) or document(s) you want removed.
2. State your claim. Explain the basis for removal (copyright ownership, privacy concern, legal order, etc.).
3. Provide contact information. Include your full name, organization (if applicable), email address, and phone number.
4. Send your request. Email your takedown notice to the address below. Include "DMCA Takedown" or "Content Removal" in the subject line.

Response Timeline

We acknowledge all takedown requests within 72 hours of receipt. Valid requests are processed and content is removed or restricted within 10 business days. We may contact you for additional information if the request is unclear or incomplete.

Good Faith Disclaimer

This project operates in good faith as a public transparency tool. All indexed content originates from official government releases. We do not make claims beyond what is supported by linked public records. Being named in these documents is not an indication of wrongdoing.