What 2.1 Million Documents Look Like
TLDR The Jeffrey Epstein document corpus contains 2,100,266 files across 12 DOJ data sets and 6 source directories, totaling approximately 331 GB. It spans...
All reports, analyses, and investigative articles.
TLDR A pipeline of 27+ Python scripts transforms 2.1 million raw government documents into a searchable PostgreSQL database with 2.38 million extracted...
TLDR The DOJ released approximately 3.5 million pages while acknowledging that more than 6 million pages were identified as potentially responsive — a 42% gap...
TLDR The Chao1 estimator — a statistical method originally designed to estimate the total number of species in an ecosystem, applied here to estimate total...
TLDR The corpus contains 224 parsed wire transfers totaling $24,059,535.45, spanning April 2004 to August 2019 across five banking institutions....
TLDR The New York Department of Financial Services fined Deutsche Bank $150 million for systemic anti-money laundering failures in the Epstein relationship. A...
TLDR A single 29-page TD Bank Suspicious Activity Report documents $47.3 million in suspicious transactions across 25 subjects from May 2015 to August 2019. It...
TLDR The Butterfly Trust, opened at Deutsche Bank in January 2014 with Indyke and Kahn as co-trustees, disbursed $2.65 million through more than 120 wire...
TLDR Benford's Law predicts that the leading digit of naturally occurring financial data follows a logarithmic distribution — digit 1 appears roughly 30% of...
TLDR OCR engines (software that converts images of text into searchable characters) produce phantom entities from blank form labels and repeated document...
TLDR A FedEx shipment from ".EFFERY EOSTIEN" to "ROBERT" at "ART CF WOMEN" in Hawaii was initially identified as linking Epstein...
TLDR Qwen2.5-VL-7B (a vision-language model — an AI system that reads images the way a human would) running on a single NVIDIA RTX 4070 with 8 GB of video...
TLDR Automated name detection software extracted 2,383,898 entities from 2.05 million documents — 57% organizations, 37% persons, 6% locations. Nearly one in...
TLDR FAA registry records document eight aircraft registered to eleven Epstein-controlled corporate entities between 1969 and 2021 (Federal Aviation...
TLDR Approximately 53 corporate entities linked to Epstein span five jurisdictions, with attorney Darren Indyke appearing as officer on more than ten of them...
TLDR The USVI Economic Development Commission granted Southern Trust Company a 90% income tax exemption in 2013 based on its claimed business of "DNA...
TLDR Custom-built parsing scripts extracted 2,894 FedEx shipments from DOJ-released invoices, spanning July 2000 through October 2005 (PAPER TRAIL Project,...
TLDR Of 148 FedEx shipments marked "Third Party," 94% still billed to Epstein's personal account -- meaning the designation identifies who initiated...
TLDR 457 Madison Avenue Floor 4, located in the landmark Villard Houses, served as the operational hub for both of Epstein's FedEx accounts (PAPER TRAIL...
TLDR Traditional text extraction failed completely on Epstein's handwritten flight logs. A vision language model -- a type of AI that interprets entire page...
TLDR A grouping algorithm that finds clusters of closely connected entities -- called Leiden community detection -- applied to 29.5 million entity...
TLDR An algorithm that finds sudden shifts in document activity patterns -- called PELT -- detected 889 verified breakpoints in document activity time series,...
TLDR A scoring method that flags documents containing unusual combinations of names -- combining two established information-theory measures called IDF and PMI...
TLDR Six parallel research agents independently verified corpus findings against public registries, FAA records, and court filings (PAPER TRAIL Project,...
TLDR Probabilistic entity resolution using a statistical matching method reduced 2.38 million raw Named Entity Recognition entries to 519,000 unified clusters...
TLDR Every analytical method in the pipeline was selected to satisfy the legal standard courts use to decide whether scientific evidence is admissible --...
TLDR Cross-domain synthesis across 232,000 events produced nine investigative leads and zero confirmed findings -- every lead fell below the 0.75 confidence...
TLDR The Epstein Files Transparency Act (P.L. 119-38) passed the House 427-1 and the Senate by unanimous consent, mandating release of all unclassified Epstein...
TLDR February 2026 was the most consequential month for Epstein accountability since the 2019 arrest. Congress compelled the first-ever deposition of a former...
TLDR Poland, France, the United Kingdom, and Lithuania opened formal criminal investigations into Epstein connections within weeks of each other in early 2026....
TLDR Twenty-nine individualized letters were generated for H.R. 4405 cosponsors and committee chairs, each containing a corpus-sourced analytical enclosure...
TLDR French entities received $1.575 million in wire transfers from Epstein-linked accounts, including $699,000 to the PIASA auction house (PAPER TRAIL...
TLDR Ghislaine Maxwell's British citizenship creates UK criminal jurisdiction across three evidence domains documented in the corpus: UBS financial accounts...
TLDR New York is the jurisdictional convergence point for the Epstein network. Deutsche Bank maintained 40-plus accounts at its New York branch, 575 Lexington...
TLDR Lithuanian prosecutors opened a pretrial investigation on February 3, 2026, based on wire transfers to "Fors projektai" (EUR 75,000) and...
TLDR A systematic search of 2.1 million documents across 12 DOJ data sets produces zero evidence supporting Pizzagate conspiracy theories (PAPER TRAIL Project,...
TLDR Every factual claim in the PAPER TRAIL documentary series is assigned one of four evidence tiers: T1 (government primary source), T2 (corpus-derived), T3...
TLDR PAPER TRAIL presents analytical methodology rather than individuals as the narrative protagonist. The machine's process of discovery -- how it extracts...
TLDR Darren K. Indyke served as Epstein's in-house attorney for more than 15 years, appeared as officer on 10-plus corporate entities, and controlled an...
TLDR Richard Kahn served as Epstein's accountant, co-executor, Butterfly Trust co-trustee, and sole authorized signer on HBRK Associates. Two months before...
TLDR Bella Klein has served as HBRK Assistant Controller since November 1999, making her one of the longest-tenured financial operatives in the Epstein...
TLDR Lesley Groff functioned as Epstein's executive assistant and scheduling gatekeeper, coordinating travel, catering, and visitor appointments documented in...
TLDR Lawrence Visoski served as Epstein's chief pilot and appeared as officer or manager on four aviation entities: JEGE Inc., JEGE LLC, Hyperion Air, and Plan...
TLDR Alfredo Rodriguez appears in FedEx records as the person who signed for packages at Financial Trust Company in St. Thomas, USVI -- including a shipment...
TLDR One of Manhattan's largest private residences — the 28,000 square-foot Herbert N. Straus House at 9 East 71st Street — passed through three corporate...
TLDR A single office address — 575 Lexington Avenue, Floor 4, New York — served as the registered location for HBRK Associates, DKI PLLC, and NES LLC,...
TLDR At least seven Epstein-linked entities registered at a single address — 6100 Red Hook Quarter B3, St. Thomas, USVI 00802. This office served as the...
TLDR Zorro Ranch in Stanley, New Mexico received approximately 29 third-party FedEx shipments billed to Epstein's account — more than any other destination....
TLDR A single Suspicious Activity Report (SAR — a filing banks are required to submit to the government when they identify potentially illegal financial...
TLDR On May 5, 2013, a single email from a Deutsche Bank executive claimed that the Head of Anti-Money Laundering (AML) Compliance had approved the Epstein...
TLDR "THE FIRST FIFTY YEARS" is a 238-page birthday book with a prologue by Ghislaine Maxwell, obtained as House Oversight estate records Request No....
TLDR The Non-Prosecution Agreement (NPA — a deal between prosecutors and a defendant in which the government agrees not to file federal charges in exchange for...
TLDR DOJ released at least eight documents containing Richard Kahn's active Deutsche Bank online banking portal credentials — user ID, account number, and...
TLDR Jeffrey Epstein's last will — 10 pages, signed August 8, 2019, two days before his death — established the 1953 Trust with Darren Indyke and Richard Kahn...
TLDR Southern Trust Company held $391.3 million in assets by 2017, was originally incorporated as "Financial Informatics Inc." for "DNA database...
TLDR Harlequin Dane LLC received $7.75 million in seven wires from Southern Financial, with Darren Indyke as sole signer. Debits went to a real estate title...
TLDR JEGE Inc. (Delaware, 2000, ceased 2018) and JEGE LLC (USVI, 2012, sold 2019) are two separate legal entities sharing the "JEGE" name. Together...
TLDR Gratitude America Ltd., a USVI-based 501(c)(3) foundation (a tax-exempt charitable organization under U.S. tax law), received $10 million from BV70 LLC —...
TLDR DKI PLLC wired $100,000 to "Black Bag Media" in Arlington, Virginia — an entity with no public records. Indyke's IOLA (Interest on Lawyer...
TLDR HBRK Associates wired $20,000 to Walch Schurti, a Liechtenstein law firm, with the memo "retainer purchase of shares Anstalt." An Anstalt is a...
TLDR Between February and May 2019, at least five Deutsche Bank accounts linked to Jeffrey Epstein were systematically closed, with wire memos explicitly...
TLDR The FedEx shipping records in the Epstein corpus end abruptly in October 2005, approximately four months before his March 2006 Palm Beach arrest. The last...
TLDR On August 8, 2019 — two days before his death — Jeffrey Epstein signed his last will, Indyke's IOLA (Interest on Lawyer Account — a special bank account...
TLDR The Non-Prosecution Agreement (NPA — a deal between prosecutors and a defendant in which the government agrees not to file federal charges in exchange for...
TLDR Deutsche Bank served as Epstein's primary financial institution from 2013 to 2018, receiving a $150 million penalty for anti-money laundering failures. TD...
TLDR Data Set 9 (531,256 files) and Data Set 11 (331,655 files) together constitute 863,000 email documents — the largest sub-corpus. DS11 was publicly...
TLDR Vision-language model processing of unredacted flight logs extracted 4,286 flights with 392 unique passenger names at zero errors. The same model on...
TLDR A fast full-text search system built into the database (known technically as a GIN-indexed tsvector) enables sub-second queries across 2.1 million...
TLDR A World War II-era statistical method for estimating how many items exist based on the serial numbers you have seen (known as the German Tank Problem) is...
TLDR A software tool that uses statistics to decide which database records refer to the same person (called Splink) merged eight or more scanning variants of...
TLDR Every analytical result in the pipeline receives a confidence grade from a military-origin system for rating how reliable a source is and how credible the...
TLDR Nearly one in four entities in the corpus (197,945 out of 821,633) appears in only one document. An entity that appears in only one document — called a...
TLDR A cleanup script (Script 19b) dissolved 8 entity groups containing 3,576 records that had been incorrectly merged because unreadable text that the...
TLDR A 190-page Ghislaine Maxwell UBS account statement (EFTA01275697.pdf) generated 190 false "Krakow" entity mentions when automated text scanning...
TLDR Les Wexner testified under deposition that he was "conned" by Epstein, despite appearing 1,000 or more times in released files, transferring a...
TLDR Bill Clinton became the first former president in American history compelled to testify under subpoena, sitting for approximately six hours before the...
TLDR Lord Peter Mandelson, former UK Cabinet Minister, was arrested February 23, 2026 by Metropolitan Police on suspicion of misconduct in public office,...
TLDR Poland established Investigation Team No. 5 to probe Epstein's trafficking and Russian intelligence connections, but the public corpus contains almost no...
TLDR The U.S. Virgin Islands Attorney General's $105 million settlement with the Epstein estate named 10 entities and exposed three distinct frauds: a 90% tax...
TLDR The entire 2.1 million document Epstein corpus was processed on a single Windows PC: an Intel i9-13950HX with 24 cores, an NVIDIA RTX 4070 with 8 GB of...
TLDR PostgreSQL 16 serves as the single master database for the entire Epstein corpus analysis. Six core tables hold 2.1 million documents, 2.38 million...
TLDR The corpus audit script (Script 27) audits the entire corpus in 17 seconds, cross-referencing 2.1 million database records against files on disk across 17...
TLDR Darren Indyke and Richard Kahn are the only two entities that appear across every analytical domain in the corpus: wire transfers, corporate...
TLDR The DOJ released documents exposing 43 or more victims' full names including 24 or more minors with home addresses, while heavily redacting perpetrator...
TLDR Senator Ron Wyden's Treasury investigation identified $1.08 billion in 4,725 or more wire transfers through Epstein accounts at JPMorgan, Bank of America,...
TLDR Darren Indyke's Interest on Lawyer Account (IOLA) at TD Bank received $14.66 million in just three wire transfers from Southern Trust Company via Charles...
TLDR The Butterfly Trust disbursed $2.65 million through 120-plus wire transfers from its Deutsche Bank checking and money market accounts, with at least 18...
TLDR A $200,000 wire from Indyke's Interest on Lawyer Account (IOLA — a special trust account attorneys use to hold client funds temporarily) to Miami criminal...
TLDR Southern Financial LLC processed $606.9 million (six hundred six million, nine hundred thousand dollars) in inflows and $194.7 million in outflows,...
TLDR Darren Indyke wired $2,839,259.73 from his personal TD Bank checking account to Horizon Title Services in West Palm Beach for "closing costs for 6030...
TLDR Wire transfers to French entities totaled $1.575 million (one million, five hundred seventy-five thousand dollars): $699,000 to PIASA auction house,...
TLDR A FedEx shipment weighing two pounds with the reference field "RADIO FENCE" cost $1,710 (one thousand, seven hundred ten dollars) to ship to Bal...
TLDR Rosemary Pitcher sent three FedEx shipments from Jet Aviation at Palm Beach International Airport to Little St. James LLC at 6100 Red Hook Quarter, St....
TLDR K L Gordon sent a half-pound envelope from Zorro Ranch in Stanley, New Mexico, to "JEANNE JAME DOE" at Financial Trust Company in St. Thomas,...
TLDR Five FedEx shipments contain JEGE-related reference field entries — including "2 ENGINE JEGE," "WON3286 JEGE" (work order), "JEGE...
TLDR A FedEx third-party shipment lists the recipient as "DR BRUCE LOSKOWTZ" in West Palm Beach. Character substitution analysis suggests...
TLDR Ghislaine Maxwell appears as both sender and recipient in FedEx records billed to Epstein's account, with shipments spanning 2001 to 2005 across Palm...
TLDR Filmmaker John Waters appears in four House Oversight documents through Peggy Siegal's Oscar Diary columns in Avenue Magazine, not through any direct...
TLDR Two observations were retracted after visual inspection revealed that scanning software (OCR, or optical character recognition — software that reads text...
TLDR Data Sets 1 through 8 were published on December 19, 2025, exactly 30 days after the Epstein Files Transparency Act became law. The initial release...
TLDR Data Set 9 is the largest single data set by file count: 531,256 email PDFs occupying 94.5 GB. It achieved a perfect disk-to-database match (zero delta),...
TLDR Data Set 10 contains approximately 950,000 page-level database records derived from 504,000 distinct multi-page PDFs, primarily Deutsche Bank financial...
TLDR Data Set 11 was publicly described by media outlets and community indexers as "financial ledgers and USVI flight manifests." It is actually...
TLDR Data Set 12 is the smallest in the Epstein corpus: approximately 150 supplemental documents occupying 121 MB. Released January 30, 2026 alongside the much...
TLDR Deputy Attorney General Todd Blanche stated DOJ identified "more than six million pages" as responsive to the Epstein Files Transparency Act but...
TLDR Virginia Giuffre's civil defamation suit against Ghislaine Maxwell, filed September 21, 2015, produced the sealed depositions and discovery documents that...
TLDR SDNY Case 19-cr-490 charged Jeffrey Epstein with sex trafficking and conspiracy following his July 6, 2019 arrest (USA v. Epstein, No. 19-cr-490, SDNY,...
TLDR The U.S. Virgin Islands Attorney General's lawsuit against the Epstein estate produced a $105 million settlement in December 2022, naming 10...
TLDR Democracy Defenders Fund filed a complaint with the Office of Inspector General (OIG — the government's internal watchdog) on February 6, 2026 and Freedom...
TLDR DOJ's January 30, 2026 release included unredacted victim names, home addresses, and nude images with visible faces. For five days, this information was...
TLDR In re Courtney Wild tested whether the 2007 Non-Prosecution Agreement (NPA — a deal where prosecutors agree not to bring charges) violated the Crime...
TLDR Ghislaine Maxwell was convicted December 29, 2021 on five of six counts including sex trafficking of a minor and sentenced to 20 years on June 28, 2022...
TLDR Southern Financial LLC wired $7.75 million into Harlequin Dane LLC, which then sent $221,050 back to Southern Financial's Deutsche Bank account labeled as...
TLDR Epstein's financial flows traversed five U.S. jurisdictions — New York, Delaware, Florida, the U.S. Virgin Islands, and Puerto Rico — plus Liechtenstein,...
TLDR An attorney servicing Epstein's accounts made 97 cash withdrawals of exactly $7,500 — Deutsche Bank's third-party withdrawal limit — at a pace of two to...
TLDR Among 2,894 FedEx shipments in the Epstein corpus, several exhibit cost-to-weight ratios that match Trade-Based Money Laundering (TBML — laundering money...
TLDR Script 18's ownership graph module constructs a directed network of 33 static edges (corporate registrations, officer appointments) plus dynamic wire flow...
TLDR Banks are required to file Suspicious Activity Reports (SARs — reports banks must file with the government when they detect potential financial crime)...
TLDR Richard Kahn made three cash withdrawals from HBRK Associates accounts totaling $62,400 in early 2019: $4,400 on February 1, $8,000 on April 22, and...
TLDR The email address jeevacation@gmail.com appears in the footer boilerplate of emails in Data Set 11, identified in document EFTA02212883 — a Lesley Groff...
TLDR Document EFTA02212885 in Data Set 11 contains a four-page email dated April 24, 2017, documenting an American Express Centurion Travel booking for a...
TLDR Both Data Set 9 (531,256 files) and Data Set 11 (331,655 files) are structured as Relativity load files — a standardized format used in litigation...
TLDR Email samples from Data Set 11 date to at least April-May 2017, placing them squarely in the post-NPA, post-conviction period when Epstein was a...
TLDR Document EFTA02212883 contains a two-page email chain from May 30, 2017, in which "jeffrey E." instructs Lesley Groff to prepare "cookies...
TLDR Rep. Thomas Massie (R-KY) served as the primary sponsor of H.R. 4405, the Epstein Files Transparency Act, and was one of only four Republicans to sign the...
TLDR Rep. Ro Khanna (D-CA) served as lead cosponsor of H.R. 4405, recruited 23 Democratic cosponsors, co-reviewed unredacted files with Massie at the DOJ...
TLDR Rep. Clay Higgins (R-LA) cast the only "No" vote against the Epstein Files Transparency Act, making the House tally 427-1 (Epstein Files...
TLDR Rep. Jamie Raskin sent a letter to Deputy AG Todd Blanche on January 31, 2026, citing $1.5 billion in suspicious bank transactions and 200,000 withheld...
TLDR Senator Ron Wyden's Treasury investigation identified $1.08 billion in 4,725 or more wire transfers through Epstein-linked accounts at JPMorgan, Bank of...
TLDR House Oversight Chairman James Comer issued subpoenas to DOJ, the Clintons, and eight former officials on August 5, 2025. When the Clintons defied the...
TLDR The PETRA Act (S.2746 — the Promoting Epstein Transparency and Releasing All records Act, a proposed law that would add enforcement teeth to the Epstein...
TLDR Rep. Lauren Boebert leaked a photograph of Hillary Clinton from a closed-door deposition to a conservative influencer, violating House rules governing...
TLDR PAPER TRAIL is a 14-episode documentary series organized into three narrative arcs: Foundation (EP01-03), Evidence Domains (EP04-10), and Synthesis...
TLDR Every claim in PAPER TRAIL is assigned one of four evidence tiers: T1 (government primary source), T2 (corpus-derived computation), T3 (journalism), or T4...
TLDR The PAPER TRAIL distribution strategy sequences international law enforcement stakeholders before domestic congressional recipients to maximize...
TLDR Seven production Scalable Vector Graphics (SVG) images — a type of image that stays sharp at any size — were created at 1920x1080 resolution with a...
TLDR All PAPER TRAIL episodes are authored as numbered markdown files (one per slide), converted to PowerPoint via Python scripts, and tracked through version...
TLDR The institutional forensics script's accountability matrix assigns risk-adjusted complicity tiers to every entity in the Epstein network using but-for...
TLDR The ownership graph built by the institutional forensics script contains exactly 33 static edges (fixed connections) representing verified corporate...
TLDR The institutional forensics script classifies institutional compliance failures into seven distinct linguistic marker categories using full-text search...
TLDR Evidence chains formalized using Wigmore and Toulmin notation — a formal method for mapping chains of evidence, showing how each piece of proof connects...
TLDR The institutional forensics script's Granger causality module — which applies a statistical test that checks whether one type of event reliably happens...
TLDR The cross-domain synthesis script generates three Analysis of Competing Hypotheses (ACH) tables — structured comparison tables that pit competing...
TLDR The three-year path from the Palm Beach Police investigation (2005) through FBI referral, the Non-Prosecution Agreement (NPA, September 24, 2007), and...
TLDR The 35 days between Jeffrey Epstein's arrest on July 6, 2019 and his death on August 10, 2019 constitute the most financially active period per day in the...
TLDR Ghislaine Maxwell's trial ran from November 29 to December 29, 2021 in the Southern District of New York (SDNY), producing a guilty verdict on five of six...
TLDR H.R. 4405, the Epstein Files Transparency Act, traveled from introduction on July 15, 2025 to presidential signature on November 19, 2025 in 127 days. The...
TLDR February 2026 saw the Epstein case breach international borders simultaneously: Lord Mandelson arrested in the UK, officials resigned in Norway, France,...