Robert Lawrence Kuhn

Named on the documents as	”Robert Lawrence Kuhn” and “Robert Kuhn” in the sender field of his own messages; “Robert L. Kuhn” in eighteen documents; “THE KUHN FOUNDATION % DR R L KUHN” on an Internal Revenue Service letter; “kuhn”, lower case and without a forename, throughout Jeffrey Epstein’s own scheduling messages12345
What a document in the release says he is	”a public intellectual, international corporate strategist and investment banker, and a renowned expert on China”, who “has worked with China’s state leaders and advised the Chinese government” for thirty years, creator and host of the public-television series Closer To Truth and of Closer to China with R.L.Kuhn on CGTN, and chairman of The Kuhn Foundation — a one-page biography, with his photograph, held in the released file6
Range of the correspondence cited here	17 August 2016 to the Epstein office schedule for June 2019, whose last entry naming him is an appointment for 13 June. Seven older sheets are also cited below and are not correspondence with him: two Internal Revenue Service letters about his charity (29 April 1999 and 17 June 2015) and five documents of 2008–2015 cited only to establish the Kahn→Kuhn extraction fault789310
Size of the file	1,219 documents read as his, established form by form below. The obvious phrase "Robert Kuhn" returns 776 of them — 64 per cent. His own middle name defeats that phrase: "Robert Lawrence Kuhn" returns 498, of which 338 are outside the base511
How much of it is his own writing	295 of the 1,219 — 24 per cent — carry his name at the head of the top-level sender field of the document’s own first sheet, under the rule published below. Those 295 resolve to between 170 and 264 distinct messages, the release holding several copies of many of them. A further 259 address him without his writing anything12
What Epstein’s money did	Epstein’s foundation Gratitude America, Ltd. sent $150,000 to The Kuhn Foundation on 1 August 2017 for the Closer To Truth series “Radical Breakthroughs”, into a sub-account Kuhn had opened for the purpose; on 15 June 2018 Epstein instructed Richard Kahn to “send 500k today” against a reminder headed “Kuhn 600k”1314
His answer in the outside record	To CNBC in December 2019 he called what had been reported about Epstein “totally reprehensible”, said he had not known of the accusations when they were introduced in 2016, and said of the unfinished series: “I wouldn’t take money from anyone else to do this and if I do finish this, then I would do it with my own money.”15
Covered elsewhere on this wiki	The December 2018 introduction Epstein made between him and Nicole Junkermann, her reply, and the 10 March 2019 conference call are set out on Nicole Junkermann’s page and are not repeated here
Not established here	That any payment reached him other than the two the documents and the outside record describe; that the “Radical Breakthroughs” episodes were made; that he knew anything of Epstein’s conduct; anything about him beyond what the sheets cited here record

Robert Lawrence Kuhn is a public-television host, author and long-standing adviser to the Chinese government who, between August 2016 and June 2019, became one of Jeffrey Epstein’s most frequent working correspondents. The relationship in the released file has three strands and they run in parallel. The first is a television project: Kuhn pitched Epstein a season of his PBS series Closer To Truth, Epstein’s foundation paid for it, and the money went into a sub-account Kuhn opened under his own charity and named for the donor. The second is China: Kuhn wrote to Epstein from Beijing about China’s leadership, appeared on CNN, BBC, CNBC and CGTN to explain it, and in 2019 ran a private inquiry in China on behalf of one of Epstein’s friends. The third is simply access: a standing appointment at Epstein’s New York house that Epstein’s office booked and rebooked for three years, the last of them fixed for 13 June 2019, three weeks before Epstein’s arrest. Nothing on this page is a finding of wrongdoing.

Finding him: a middle name, a surname, and somebody else’s surname

This subject is the clearest case on this wiki of a name that the obvious search cannot reach, and of a search that reaches the wrong man.

The phrase problem. The full-text index tokenises, and an exact-phrase query requires adjacency. So "Robert Kuhn" cannot match a sheet that prints Robert Lawrence Kuhn. Measured on 2026-09-12 against the corpus full-text index, which covers the EFTA, HOUSE_OVERSIGHT and DOJ-OGR productions together, and counting distinct documents: "Robert Kuhn" returns 776; "Robert Lawrence Kuhn" returns 498, of which 338 are outside the 776; the union of the two is 1,114. A page built on the base phrase alone would have seen 70 per cent of that union and none of the gap.⁵

The damaged forms. The desk’s variant generator, ocr_variants.py, produces one candidate for "Robert Kuhn" ("R0bert Kuhn", zero documents) and none at all for "Kuhn". Every damaged form that actually exists in this release was found by reading, not by generating: "Roberi Kuhn" (3), "Roben Kuhn" (4), "Robed Kuhn" (1), "Rober1 Kuhn" (1), "Robert Kuhi" (1), "Robert Kuhr" (1), "Robert Lawrence Kuhr" (4), "Robert Laurence Kuhn" (1), "Robert L. Kuhn" (18), "Robert Lawrence runn" (1), "rlkuhn" (16), "R. L. Kuhn" (10). Adding every one of those to the union brings it to 1,129; the 15 documents they add were each opened and each is this subject.⁵¹⁶

The surname, and what it drags in. Bare "Kuhn" returns 1,279 documents, which is 165 more than the two-name union. Those 165 were read one by one. They do not reconcile into a residue; they partition, and the two biggest buckets point in opposite directions:

what the 165 are	documents
this subject, under the bare surname or a damaged forename	102
Richard Kahn, whose surname the extraction renders “Kuhn”	7
Thomas Kuhn, the philosopher of science — 6 of them reached by the phrase "Thomas Kuhn", 6 only in other forms (“T. S. Kuhn”, “Kuhn TS”, “Kuhn describes”, “Kuhn-like”)	12
Thomas Kuhn, president of the Edison Electric Institute — a different man of the same name	1
academic citations to J. Kuhn, D. Kuhn, R. M. Kuhn, A. Kuhn and Markus Kuhn	13
the case citation Litton Indus. v. Lehman Bros. Kuhn Loeb Inc.	4
twelve other individuals named Kuhn — a hedge-fund partner, a tax attorney, a Wells Fargo vice-president, an FBI records official, a jazz pianist and others	15
extraction noise: a page of cell codes, a call-detail key, an airline field, and the Kulin nation rendered “Kuhn nation”	11

102 + 7 + 12 + 1 + 13 + 4 + 15 + 11 = 165.¹¹

The 102 matter more than the arithmetic. They are not marginal mentions. They are Epstein’s own shorthand (“calls kuhn”, “tell kuhn 1130”, “Kuhn june 2”, “sat kuhn”, “conference call in for kuhn ?”, “reviewd today with kuhn your situation”, “kuhn is in china”), and a run of office day-sheets on which the extraction damaged the forename. The most operational traffic about this man is precisely the traffic his full name cannot reach.¹¹¹⁷

The seven are a fault running the other way, and it is image-verified. On EFTA02216282, a Lesley Groff schedule of 11 July 2017, the text layer reads “TBD Rich Kuhn”. The sheet prints “TBD Rich Kahn”. On EFTA01579234, a 2009 wire memorandum, the text layer reads “Please call Rich Kuhn with the Fed Reference number”; the sheet prints “Please call Rich Kahn”. On EFTA00354583, a January 2015 distribution list, the text layer reads “Richurd Kuhn”; the sheet prints “Richard Kahn”. All three were rendered locally at 300 dpi and read, and the other four in the bucket were rendered at 600 dpi and read for this page. One of those four defeats the rule that settled the other six: its page image is itself typeset output of an OCR pass, so the image prints Rich Kuhn as cleanly as a document would, and only the clean scan of the identical transmission under another Bates shows Rich Kahn. Kahn → Kuhn is a real extraction fault, and it inserts a different person into this subject’s surname population. A count keyed on the surname alone would have credited Epstein’s accountant’s wire instructions to a television producer.¹⁰

Taken together, the documents read as his are 1,219: the 1,129 reachable by some form of his name, plus the 90 of the 102 bare-surname documents that no name form reaches. The base phrase reaches 63.7 per cent of that.¹¹

How much of this file is his own writing

A count of documents naming a person is not a count of what that person did. The rule below is stated completely enough to be re-implemented.

The rule. For each document, take its own first sheet (the row with the lowest page_number for that identifier) and the first twelve non-blank lines of it. Find the first line matching a sender label anchored at line start, over the damaged forms of From: the extraction produces (Fran, Front, Frorn, Frnm, Fram, Fnxn, Fromm, Fm, Frm, Fron, Erom, Frcrn, Ftom, Fiom, Prom, Sender). The sender value is the remainder of that line if it is non-empty; otherwise the next non-blank line, unless that line is itself a header label or begins with a bare header word, in which case the sender value is absent and the document is declined. Only the first sender label is used, so a quoted header lower on the sheet can never supply the sender.

The result, over the 1,219: 891 documents yield a top-level sender; 202 carry a sender label whose value is absent, barred or blank and are declined; 126 carry no sender label at all.

• 295 documents, 24 per cent, carry his name in that top-level sender value, and in all 295 his forename stands at the head of it.
• 259 documents address him in a To: or Cc: value without his name appearing in the sender value.
• The rest name him in a body, a schedule, an attachment list or a press capture.

The figure is stable, and it is not the tool’s figure. Re-running the same rule at windows of 8, 12, 20 and 30 non-blank lines gives 294, 295, 295 and 295 over the full population. The desk’s act_rate.py, whose --selftest printed PASS before every figure here was taken, returns 280 of 776 on the base phrase and 378 of 1,114 on the two-name union. Over the same union this rule returns 294. The gap of 84 is the tool’s known inline-branch fault, and the fault is more general than “it joins labels”. Reading the source: one test (act_rate.py:127, if any(v for _,_,v in labels)) routes a whole sheet down the inline branch as soon as any label in the first 2,500 characters carries an inline value. The inline branch therefore preempts the top-level header, and a top-level header that is stacked, or whose sender value stands under a redaction bar, contributes nothing, so the sender is taken from whatever is quoted below. Sixty-four base-population documents are scored as his by the tool and not by this rule. On thirty of them the tool has joined two or more sender labels, and on twenty-nine of the thirty the first sender label on the sheet is somebody else: Lesley Groff, Richard Kahn or Epstein. On the other thirty-four no joining happens at all: the topmost sender label carries no inline value on every one of the thirty-four, so the only inline sender value anywhere on the sheet is a quoted one. On this subject the tool overstates; on people/reid-hoffman the same line of code made it understate by 122 documents on that page’s own measurement (239 against 361). The sign is set by the population’s layout, not by the tool: where the subject sits at the top of a stacked header and others are quoted inline below it understates, and where others sit at the top under a bar and the subject is quoted inline below (this population, an office that forwarded Kuhn’s messages constantly), it overstates. Neither number should be quoted without its rule, and this page quotes 295 and has just given the rule.¹²

A second, looser rule is stated because the first one has a price. The strict rule declines the stacked header layout (all labels, then all values) whenever the sender label carries no inline value. Mapping those positionally instead, refusing when there are fewer values than labels or when the value landing on a Sent/Date label is not date-shaped, gives 305 rather than 295. The published figure is 295 and the span under the two rules is 295–305; the choice is named rather than hidden.¹²

295 documents are not 295 messages. Of the 294 on the two-name union, 200 yield a parsable top-level timestamp and those 200 resolve to 170 distinct timestamps; thirty timestamps carry more than one document, covering sixty documents between them. With the 94 unparsed added at worst one-for-one, his own outbound traffic in this release is between 170 and 264 distinct messages, running from 21 August 2016 to 30 May 2019. By year, the 170 dated distinct messages fall 24 in 2016, 74 in 2017, 35 in 2018 and 37 in 2019.¹²

He is a broadcaster, and the standard press test misses him almost entirely. act_rate.py reports zero press documents on the base phrase and zero on the union. That zero is an artefact of its window: the press arm reads only the first 1,200 characters of the first sheet and stops. All 61 documents the tool files as “header-less, authored” on the union were read. Thirty-six of them are press: CNN, CNN International, CNN’s Amanpour, Fareed Zakaria GPS, BBC World News, Bloomberg Television, CNBC and CGTN transcripts of his own appearances, together with China Daily, South China Morning Post, Shanghai Daily and International Herald Tribune columns and profiles under his byline. Two are copies of his biography sheet; twelve are Epstein office schedules; one is a contact card; two are Skype call logs; and only eight carry his own writing; on two of those eight Epstein’s own words stand at the head of the sheet and Kuhn’s message is quoted below, so “carry” is the right verb for them and “authored” is not. On the base phrase the same column is 19 documents, of which two carry his writing and one is press. A zero from that arm means “no press marker in the first 1,200 characters”, not “no press”, and the caption telling you to add the header-less column only if you have read some is the only thing standing between it and a false act rate.¹⁸

The introduction, August 2016

The file opens with Epstein asking to be introduced.

On 17 August 2016 Gino Yu, then corresponding with Epstein, sent him a Closer To Truth episode link under the subject line “Kuhn Interview… Divine intervention.” The body of that message is the link and nothing else.⁷ The next day Epstein wrote back to Yu: “can you introduce me to kuhn by skype.”¹⁹ On 20 August Yu wrote to Kuhn: “I’d like to introduce you to Jeffrey Epstein who I was going to see if you could meet with while I was in NYC. He has an interest in AI and consciousness and I showed him your Closer to Truth site.”²⁰

Kuhn answered both men early on 21 August 2016, and the message is the earliest document in this file that he wrote:

Hi Gino: Appreciate the intro. Lots of common pursuits!

Hi Jeffrey: Great to Skype, meet in NYC. I’m current shooting Closer To Truth at an FQXi conference (in Banff) on foundations of physics and quantum cosmology. (Just with Alan Guth, an old friend, for on-camera discussions - 5th time.)

He attached his essay “Virtual Immortality” from Skeptic magazine and said he would be in New York for the first half of October.²⁰ Later the same day the two men were already trading names. Epstein wrote that he had spoken to Seth Lloyd; Kuhn sent back the list of on-camera topics he had recorded with Lloyd the day before and, an hour later, the day’s interview list: “David Wallace, Tegmark (I’ll say ‘Hi’) Bernard Carr, Paul Davies (one of Closer To Truth’s core regulars).”²¹

Epstein chased. On 20 August he told Yu, “did not hear back from kuhn. send him, a slyep request”; on 22 August Yu reported back, “Kuhn responded by mail.”²²

The meeting happened on 4 October 2016 at Epstein’s house. A message from Epstein’s office, whose sender field is barred and which is signed “Assistant to Jeffrey Epstein”, told him: “Reconfirming you will come see Jeffrey today at 2:30pm.” The next line of the sheet is the address, “9 East 71st Street between 5th and Madison”, with no full stop after it.²³ Two days earlier Epstein had set out his week to his office in a single line of lower-case shorthand: “jabor monday afternoon. leon tues fenwick, kuhn, churckin thiel.summers dinner woody thurs rothchild tom barrack, slayton karp blaine —sunday depak gromov.” That line survives in fifteen copies, and eleven of them are reachable by neither form of his name.⁴

On 6 October Kuhn wrote the first of the documents that define the relationship:

Jeffrey:

Great to meet, engage on multiple levels.

We’ve posted three Closer To Truth episodes for you (Time, Information, Minsky Tribute). Links follow.

Beneath it the sheet sets out a table of episodes and their contributors (Tegmark, Penrose, Chalmers, Dennett, Wilczek, Rees) under the heading “OTHER EPISODES WE CAN POST”, with the note “Let me know if you would like any of these."¹

"Shocked in Beijing”, 9 November 2016

The day after the United States presidential election, Kuhn wrote to Epstein under that subject line, marked high importance. The message opens:

You are a prophet…

and continues:

I was just with senior leaders in Beijing…

He then told Epstein he would be discussing China–United States relations under the incoming president with the former Chinese ambassador to the United States on CCTV News, on the programme Closer To China, broadcast in China and internationally, and gave two New York broadcast times, of which the sheet prints one intact, Sunday at 8:30 am. The Saturday time prints ”=:30 pm”, the hour eaten by the transport damage described in the footnote, and is not reconstructed here.

The paragraph that follows runs unbroken on the sheet from “We’re playing in new territory here.” through “To be discussed privately.” and on into his travel plans; its last words are his return to Los Angeles on 17 November and New York on 24 December, “after my =ext trip to Beijing”, in which next has lost the same character to the same damage. The offer and the reticence stand in the middle of a paragraph about diary dates, not at the end of one.²⁴

What the contributions he offered were to be is not stated on the sheet, and no document read for this page says. The sentence is quoted here because it is his, it is dated, and it was written to Epstein.

The proposal, the charity check, and the money

On 11 January 2017 Kuhn sent a formal pitch:

Dear Jeffrey:

Attached, as promised, is our proposal for Closer To Truth to =roduce the definitive television series on “The Science of =leep and Dreams” - 13 episodes, a full season, on ~200 PBS =tations, plus extensive web videos.

That is what the sheet prints, and it is quoted here with its damage. The sheet carries quoted-printable damage which eats a single character at three points in that sentence: read produce, Sleep and stations. ~200 is not damaged: the sheet prints a tilde, “approximately 200 PBS stations”, and it is the text layer, not the document, that turns it into a quotation mark. This page prints the damage and glosses it rather than repairing it silently, because a silent repair cannot be told from a transcription error by anyone checking the page against the image.²

Three attachments went with it, a twelve-page proposal, a line-item budget and a weekly schedule, and the proposal’s filename names the funder. It prints on the second sheet, with the same damage and an en dash: Closer To Truth - The Science of Sleep & =reams – CTT TV Episodes & Web Videos - Jeffrey =pstein VI Foundation - 1.10.17.docx, in which Dreams and Epstein have each lost a character. Eleven days later, in the same thread, he added a line of his own about the election:

I forgot to congratulate you on winning =our loooooong-odds bet on Trump with Woody. Everyone I know, including =e, would have bet against you. Quite extraordinary.

The same damage eats the y of your and the m of me; loooooong, with six os, is Kuhn’s own spelling and prints that way on the sheet.²

Epstein pressed him on the substance. On 15 February 2017 Kuhn answered at length, opening “Pardon the formality below - I want to approach your very legitimate questions / concerns in a serious manner”, and setting out, with Boston University’s Professor Patrick McNamara as his source, why sleep and dream research justified a series and why 2017 was the moment for it. Epstein’s reply that afternoon was seven words: “is it a charity that received the funding?”²⁵

The answer took five months to become a wire. On 1 August 2017 Kuhn wrote to Epstein:

I opened a sub-account under The Kuhn Foundation 501(c)(3), which is dedicated to our project, to facilitate monitoring use of funds - ‘Kuhn Foundation CTT JE – Radical Breakthroughs’

Epstein forwarded it the same morning to his accountant Richard Kahn with a three-word instruction: “150k from gratitude”. The attachment is a Citibank wire-instruction sheet headed “Kuhn Foundation CTT JE – Radical Breakthroughs / Incoming Wire Instructions”, in favour of a Morgan Stanley Smith Barney account; the bank routing values on the sheet read for this page are covered by black bars and are not reproduced here.¹³

Kahn did his own check and could not make it work. He replied to Epstein the same afternoon: “i cannot figure out which is robert’s foundation when checking 501 (c)3 status on IRS website / can i reach out to him to verify / sending money now but want backup for our files”. Beneath his message the sheet carries his screen capture of the Internal Revenue Service’s Select Check results for the word “Kuhn”: sixteen organisations, including “Kuhn Foundation”, “Kuhn Family Foundation”, “The Nussbaum Kuhn Foundation” and “Wolf Kuhn Foundation”. Kahn was sending the money before he had resolved which charity he was sending it to. Kuhn supplied the paperwork that evening: two IRS determination letters, “(1999) for ‘The Kuhn Family Foundation’ and change of name to ‘The Kuhn Foundation’ (2000)” and an “Updated IRS letter (2015)”.²⁶

Both letters are in the release. The 1999 letter is addressed to THE KUHN FAMILY FOUNDATION, c/o Geneva Companies, and recognises it as exempt under section 501(c)(3) and as a private operating foundation under section 4942(j)(3). The 2015 letter is addressed to “THE KUHN FOUNDATION / % DR R L KUHN THE GENEVA COMPANIES / 357 WAVERLY DR / PASADENA CA 91105” and confirms the same status.⁹³

Ten months later Epstein went back for more. On 14 June 2018 he wrote to Kahn: “robert kuhn we need to send 600k from foundation . remind me torow”. Kahn answered the next morning under the subject “reminder Kuhn 600k”, in four lines which print on the sheet with the same character-eating damage, quoted here as they print, with an oblique for each line break: “yes we sent 150k on 8/1/2017 =rom gratitude / please confirm i will send 600k today / any pape=work we need darren to draft? / please advise”. Read from gratitude and paperwork; the damage is on the document and is not repaired here. Epstein first wrote back, at 9.52 that morning, on printed p.2 of the same run: “closer to truth, we=had wired 150 k previosulsy”. Read we had; the misspelling of previously is Epstein’s own, and no undamaged copy of that sentence exists in the release.¹⁴ Then, at 2.44 p.m. on 15 June 2018, he sent the operative instruction:

send 500k today

That is the whole body of the message.¹⁴²⁷

The outside record closes the loop. CNBC reported on 12 December 2019, from Gratitude America’s Form 990, that the foundation sent $500,000 to the Kuhn Foundation in 2018 and $150,000 in 2017, and that Kuhn said Epstein had agreed to fund the project in its entirety at $1.4 million. The episodes, CNBC reported, were never completed.¹⁵

”Radical Breakthroughs”

The series the money was for changed its name and its subject. By mid-2017 the working title is “Radical Breakthroughs”, and the correspondence is a working editorial correspondence, with Epstein commissioning and Kuhn revising.

Epstein’s editorial notes were blunt. On 28 December 2017 he wrote: “great work , but needs some editing, did not like multi verse and sensience. seems more blahblah”. Kuhn’s answer the next day opens “`Blahblahs’ forbidden in ‘Radical Breakthroughs’. So we need two replacements (for Multiverse as Science and Rare Sentience)”, and the message that follows sets out a revised episode list: six mathematics episodes where there had been fewer, a new episode on the replication crisis (“Deep problems of replicability in biomedical and psychological research - the scandal in recent years”), and a note that “Assuming we expand math episodes, we will need more mathematicians.”²⁸²⁹

Between those exchanges Kuhn fed Epstein material. On 15 July 2017, under the subject “Math of Fake News”, he sent a Scientific American piece with the note: “Great story for one of our math episodes - popular topic, serious substance”.³⁰ On 24 July 2017, under “Music/Brain”, he sent a Nature essay: “Essay / book review ‘How music meets mind’ in Nature - good leads.”³¹ In September 2018, when Epstein wrote that he wanted to fund work on music, writing “does its production give one a window on the creative mechanism”, Kuhn folded it into the series within the hour, answering Epstein and the UCLA neurologist Mark Tramo together: “Closer To Truth would love to support as you develop (TV and web)”.³²

By November 2018 the project had a contributor spreadsheet. Kuhn wrote, copying his director Peter Getzels: “Breakthroughs invites to contributors going very well - attached is our working spreadsheet. When you’re in town, Peter and I can come over, discuss (briefly) vision, production, characters - get your feedback.”³³ In February 2019 he wrote to the mathematician Misha Gromov, copying Epstein, in terms that put Epstein inside the production:

As you know, Jeffrey and Closer To Truth are collaborating on a major series for PBS on “Breakthroughs” in mathematics and other areas (e.g., biology, music) - and we surely want to feature you in one or more of the mathematics episodes.

He told Gromov filming would be in New York in April and probably June.³⁴ On 5 April 2019 he sent Epstein a reading pack on the mathematician Gregory Chaitin, ten attachments, with a summary of Chaitin’s work on the halting probability and the closing line “See you at 11 am tomorrow/Saturday.”³⁵ Lesley Groff’s note to Epstein that same afternoon asked: “Would you like lunch served tomorrow at noon with Kuhn, Chaitin family and Peter Getzel?” Epstein’s answer was “yes”.³⁶

China, and the inquiry into a business partner

The China strand is continuous. Kuhn wrote from Beijing, sent Epstein his broadcast links, and in the last year of the file did something more specific.

On 13 December 2018 Epstein introduced him to Nicole Junkermann, who was in dispute with a Chinese business partner. That introduction, her reply and the March 2019 conference call are set out on her page and are not repeated here. What belongs on this page is what Kuhn then did, which he described to Epstein himself.

On 21 January 2019 he sent an assessment of the man on the other side of the dispute:

Update: Wu (partner) is an astute business guy from Wenzhou (Zhejiang Province), a city known for savvy, hyper-aggressive, shady entrepreneurs. He does not have serious political connections, especially not in Shanghai - which for us is good news. The informal assessment is that he taking advantage of a foreign partner simply because he can.

In the same message he identified his own instrument: “my Chinese partner in Beijing, Adam Zhu - in the sports business as I think I mentioned (he was the CEO of IMG China) - seems to have uncovered a mutual friend that Adam and Wu have in common (a major Chinese sports entrepreneur).” He closed by asking Epstein: “Should you or I update Nicole?” Epstein’s reply was “Neither,?lets wait until something solid”.³⁷

What Epstein did with the assessment is on another page of this wiki, and the two sheets bear directly on a question that page leaves open. The same night, at 00:59 on 22 January 2019, having just told Kuhn to wait, Epstein wrote to Nicole Junkermann under the subject line “do not pass on”: “Wu from Wenzhou (Zhejiang Province). super shady, no real high level party strength. but clever,sneaky. etc.” Every element of that sentence stands in the message Kuhn had sent him that evening: the city and the province in the same parenthesis, the word shady, and the finding that Wu has no serious political connections, compressed into Epstein’s own shorthand. Nicole Junkermann’s page quotes the description and states that where Epstein got it is not established by any document read for that page. The sequence is a documented fact: Kuhn’s assessment reached Epstein first, and the same three elements went on to Junkermann the same night. What the documents do not establish is that Epstein had no other source, and nothing here asserts that he did not.³⁸

Ten days later, on 31 January 2019, Kuhn reported again:

Adam will meet with Wu through the mutual friend - naturally, as he is CEO of sports events company and had minor prior relations. As noted, Wu has no strong political connections, which is very helpful - so we will try to find person/people who affect him or business. That’s China. If progress, I will meet on next trip early March.

The same message gives a valuation view, “Valuation is down - last 2 years poor performance, hurt by e-commerce Alibaba and JD.Com”, then two CNBC links from that day, then the production schedule: “NY, LA, SF shoots being set - Peter just spent a week in LA and SF talking to contributors/interviewees and scouting locations.”³⁷

The television series and the private inquiry ran down the same wire, in the same messages.

Epstein’s own side of this is in his shorthand. On 9 January 2019 he told Junkermann, “reviewd today with kuhn your situation”; on 4 March, “making some progres in china. i would like you to talk directly to kuhn pick a time”; on 8 March, “kuhn is in china”; on 10 March, to Darren Indyke, “conference call in for kuhn ?”³⁹ His estimate of what Kuhn could do was not unlimited: writing to Junkermann on 4 January 2019 he had said the matter needed “heavy players” and added, “not likely kuhn.”⁴⁰

The standing appointment

Beneath the projects there is a simpler record: three years of Epstein’s office booking him in.

Epstein’s assistants wrote to him as a fixture. “Mr. Kuhn, Lesley is out of the office until next week and I am filling in for her. Would you please let me know your availability to meet with Mr. Epstein between April 12-14?”, wrote Bebe Avdiu of Darren K. Indyke PLLC, 10 April 2017.⁴¹ On 5 October 2016, the day after the first meeting, the office wrote again: “Hello Robert. Jeffrey is asking if you would like to join him for dinner this Thursday evening at his home…Woody Allen and Soon Yi will be coming along as well.” Kuhn answered the same morning, and his answer was a question of his own: whether he might bring his wife.⁴² On 15 June 2017 the house manager Merwin Dela Cruz wrote to Epstein four words: “Sir, Mr Kuhn is here”.⁴³

The office schedules carry him from October 2016 to June 2019 (“4:00pm Appt w/Robert Kuhn”, “9:00am BREAKFAST w/Robert Kuhn”, “TBD Robert Kuhn?”, “Reminder: Robert Kuhn to depart NY around 10th”, “Reminder: Robert Kuhn is hoping to see you Friday”), and on the three EFTA sheets the extraction damages the forename the sheet prints correctly, which is why the surname search finds them and the name search does not.⁴⁴

The last exchange in the file is of 30 May 2019. Kuhn wrote to Lesley Groff: “Hi Leslie: Greetings from Beijing! I look forward to meeting with Jeffrey, as always. I return to NY on June 2 just in time for our second NY Breakthroughs shoot June 3- 5 (always intense).” Groff replied that Epstein would be in New York only on 2 June and proposed the 13th; Kuhn answered, “Great; 4:00 pm, June 13. Looking forward / Robert”.⁴⁵

Epstein’s office entered it. The schedule for June 2019 carries, under “Sun. June 2. 2019 NY”, the line “Reminder: Robert Kuhn to interview in NY June 2-5 with Gintis, Kaku, others”; and eleven days later, under “Thurs. June 13, 2019 NY”, the line “4:00pm Appt w/Robert Kuhn”, the hour Kuhn had proposed a fortnight earlier. The same day’s reminder line on the same sheet reads “Rich Kahn on holiday June 13/14, 17 return on Tues. June 18”, with the accountant’s surname printed correctly, which is the cleanest single demonstration on this page that the two names are distinct and that only the extraction confuses them.⁸ Epstein was arrested on 6 July 2019.

A schedule records that a meeting was entered, not that it happened. No document read for this page records whether the 13 June appointment was kept.

What the outside record says, and his answer

Kuhn’s own account is on the record and belongs beside the documents. CNBC reported on 12 December 2019 that Gratitude America’s most recent Form 990 showed a $500,000 payment to the Kuhn Foundation in 2018, the largest donation Epstein made in the year before his arrest, and a $150,000 payment in 2017; that Kuhn said Epstein had committed to funding the project in full at $1.4 million; and that the mathematics episodes were never completed. Kuhn told the network: “The obvious answer is that it’s totally reprehensible what’s been reported and it’s an example of how corruption can be used in human control.” On finishing the series he said: “I wouldn’t take money from anyone else to do this and if I do finish this, then I would do it with my own money,” and “I feel an obligation to finish it and wouldn’t ask anyone to fund it with this baggage.” He said he had no knowledge of the accusations against Epstein when the two were introduced in 2016.¹⁵

That is an account of what he knew, and this page does not dispute it. The documents establish what he wrote, when, and what was paid. They do not establish what he knew.

What this page does not establish

The searches that produced the population above are printed in the footnotes with their forms, their dates and their returns. Three limits are worth stating plainly.

The 90 bare-surname documents are his by reading, not by query. Each was opened and the line carrying the string was read. A reader who re-runs "Kuhn" and subtracts the name union will get 165 documents, not 102; the assignment above is a judgement about each one, and the table is there so it can be overruled.

The payment record here is an instruction record. The corpus documents show Epstein instructing $150,000 and then $500,000; they do not contain a bank confirmation for either. The confirmation that the money moved comes from the outside record, Gratitude America’s Form 990 as reported by CNBC, and is cited as that and not as a document in hand.¹⁵

Nothing here bears on the episodes’ fate. A document that disconfirmed this would have to say, in its own words, that the money was going back or the work was stopping. Twelve such forms were searched against the whole index and none returns a single document inside this subject’s population: "return the funds" (33 documents corpus-wide, 0 here), "return the money" (38, 0), "refund" (9,742, 0), "repay" (1,819, 0), "repaid" (1,321, 0), "returning the" (399, 0), "give back the" (47, 0), "not proceed" (241, 0), "wind down" (241, 0), "cancel the series" (0), "cancelled the series" (0), "terminate the project" (0). Nor does the population itself run on: of the 1,219 documents, 1,177 yield a year in the first fourteen non-blank lines of their first sheet, taking the first four-digit token of the form 19xx or 20xx to appear there, and not one of those years is later than 2019 (2016: 192, 2017: 473, 2018: 181, 2019: 314, and 17 earlier, which sum to the 1,177). The last dated entry is the office schedule for June 2019. No document read for this page says what became of the series, and this page does not assert that one exists.⁵⁴⁶

Sources

Every quotation below is a single contiguous span from one sheet. Where a sheet was rendered locally from the released PDF and read at the image, the footnote says so and gives the resolution; where it was not, the footnote says the text layer only. Every sheet from which this page takes a displayed quotation has been read at the image, and where the sheet is damaged the damage is printed and glossed rather than repaired: a repair made silently cannot be told, by a reader checking this page against the document, from a transcription error. Where a value is covered by a black bar, nothing is supplied in its place, and nothing barred on one copy is completed from another. Bank routing values, identification numbers, message addresses and telephone lines are not reproduced.

Related

No link below asserts a relationship beyond what the cited sheet shows.

• The page that already named him, and whose account this page follows rather than repeats: Nicole Junkermann. The 13 December 2018 introduction Epstein made, her reply of 17 December, Kuhn’s of 14 December, and the conference call of 10 March 2019 are all set out there. This page carries the other half: what Kuhn did in China once introduced.
• The money: Gratitude America, Ltd., the Epstein foundation named in the instruction “150k from gratitude” and identified by CNBC as the payer of both amounts; Richard Kahn, who executed both instructions and whose own surname the extraction turns into this subject’s; HBRK Associates Inc., the firm in Kahn’s signature block; Darren Indyke, named in Kahn’s question “any pape=work we need darren to draft?”, printed here with the sheet’s damage, as above; read paperwork; and the recipient of the 10 March 2019 conference-call note.
• People named on sheets cited here: Jeffrey Epstein (the correspondent throughout); Lesley Groff (copied on the first episode list, and the fixer of the last meeting); Bella Klein and Leon Black and Kathryn Ruemmler (named on the schedules that also carry him); Steve Bannon (named beside him in Epstein’s shorthand of 4 January 2019); Marvin Minsky (the subject of one of the three episodes Kuhn posted for Epstein in October 2016).
• The surname trap: Richard Kahn is a different person, and seven documents in the bare-surname population belong to him because the extraction renders Kahn as Kuhn. All seven were checked at the image.
• Places: 9 East 71st Street, the address at which his October 2016 appointment was reconfirmed and at which the house manager announced him in June 2017.
• Topics: Calendars and daily schedules, whose own text notes that appointment records establish only that meetings were scheduled or planned; Epstein office monthly schedules, the series to which the June 2019 sheet belongs; Epstein and academic institutions and Epstein network: academia, for the wider pattern of science funding of which this is one instance.

Footnotes

1. Robert Lawrence Kuhn to Epstein, copy to Lesley Groff, Subject “Kuhn - Closer To Truth episodes with Links”, “Date: Thu, 06 Oct 2016 00:50:51 +0000”. Read at the image at 300 dpi by the writer on 2026-09-12. His own and Groff’s addresses are covered by black bars. The sheet prints “Robert Lawrence Kuhn”; the extraction renders it “Robert Laurence Kuhn”, which is why this document is reachable by neither of the two name phrases. https://epstein-data.com/EFTA00317670 p.1. ↩ ↩²

2. Robert Kuhn to Epstein, Subject “Re: Kuhn”, “Sent: Sunday, January 22, 2017 9:49 PM”, carrying his messages of 22 January and of “Jan 11, 2017, at 2:51 AM”. Both printed sheets were read at the image at 600 dpi on 2026-09-12. On the image the sender line prints From: Robert Kuhn followed by a black bar over the address, and the quoted attribution of the 11 January message prints =obert L. Kuh before a second bar; the extraction renders the sender value “Robert Kuhi”, which is why neither name phrase reaches this document. The sheet carries quoted-printable damage that eats one character at fifteen points on p.1, five of them inside the two passages quoted above and printed there as they stand. An earlier draft of this page repaired those five silently — printing produce, Sleep, stations, your and me — printed "200 where the sheet prints ~200, printed l000000ng with digit zeros where the sheet prints loooooong with six letter os, and carried a footnote certifying that every span quoted above printed undamaged, on a sheet that footnote also said had never been rendered. Both quotations above are now transcribed from the image. The three attachment filenames print on p.2 with the same damage and with an en dash, not the em dash an earlier draft of this page gave them; the first is quoted in the body as it prints, and the other two are a series budget and a series schedule-timeline. https://epstein-data.com/EFTA02662499 p.1; attachment list at p.2. The run is 2 sheets and both were read: every passage quoted above stands on printed p.1 except the attachment filename, which stands on p.2 with the closing lines of the 11 January message. Because this footnote names both sheets of a two-sheet run, a mechanical test of its pins cannot fail them; the split is stated here so a reader can check it by eye. ↩ ↩² ↩³

3. Internal Revenue Service letter LTR 4168C, “June 17, 2015”, addressed to “THE KUHN FOUNDATION / % DR R L KUHN THE GENEVA COMPANIES / 357 WAVERLY DR / PASADENA CA 91105”, confirming recognition as exempt under section 501(c)(3) “in a determination letter issued in April 1999” and as “a private operating foundation described in section 4942(j)(3) of the Code”. Read at the image at 300 dpi by the writer on 2026-09-12 and again at 600 dpi on the same date. Identification-number values on this sheet are not reproduced here, and none is completed from any other source. https://epstein-data.com/EFTA00596833 p.1. The run is 2 sheets. ↩ ↩² ↩³

4. Epstein’s week list of 2 October 2016, 2:45 AM, beginning “jabor monday afternoon.” The phrase "jabor monday afternoon" returns 15 documents, of which 4 also carry a full form of his name and 11 do not: https://epstein-data.com/EFTA00317799, https://epstein-data.com/EFTA00317811, https://epstein-data.com/EFTA00317890, https://epstein-data.com/EFTA00816368, https://epstein-data.com/EFTA00816475, https://epstein-data.com/EFTA00816481, https://epstein-data.com/EFTA02040399, https://epstein-data.com/EFTA02040603, https://epstein-data.com/EFTA02040850, https://epstein-data.com/EFTA02040919, https://epstein-data.com/EFTA02041042 — each p.1, each read in the text layer. The four inside the name population are https://epstein-data.com/EFTA00317805, https://epstein-data.com/EFTA00816350, https://epstein-data.com/EFTA00816355 and https://epstein-data.com/EFTA02041156. ↩ ↩²

5. Every count on this page is of distinct documents, not pages, taken on 2026-09-12 with exact-phrase FTS5 queries against the corpus full-text index, which covers the EFTA, HOUSE_OVERSIGHT and DOJ-OGR productions together. LIKE was not used anywhere. The forms tried and what each returned: "Robert Kuhn" 776 · "Robert Lawrence Kuhn" 498 (338 outside the base) · union of the two 1,114 · "Robert L. Kuhn" 18 (2 outside the union) · "Robert Lawrence Kuhr" 4 (2) · "Roberi Kuhn" 3 (3) · "Roben Kuhn" 4 (1) · "Robed Kuhn" 1 (1) · "Rober1 Kuhn" 1 (1) · "Robert Kuhi" 1 (1) · "Robert Kuhr" 1 (1) · "Robert Laurence Kuhn" 1 (1) · "Robert Lawrence runn" 1 (1) · "rlkuhn" 16 (1) · "R L Kuhn" 10 (1) · "R. L. Kuhn" 10 (1) · "Kuhn Robert Lawrence" 3 (0) · "Dr Kuhn" 6 (0) · "Dr. Kuhn" 6 (0). Union of all forms above: 1,129. Forms tried that returned nothing: "Bob Kuhn", "Rob Kuhn", "Robert Kuhn Jr", "Dr. Robert Kuhn", "R0bert Kuhn", "Robert 1awrence Kuhn", "Rcbert Kuhn", "Robert Knhn", "Robert Kulm", "Robert Kuhne", "Roberto Kuhn", "Robert Lawrence Kuhi", "Robert Lawrence Knhn". Surrounding counts: bare "Kuhn" 1,279 · "Thomas Kuhn" 7, with zero overlap with "Robert Kuhn" · "Kuhne" 7 · "Kulm" 2 · "Knhn" 2 — and none of those last three can be a member of the bare-"Kuhn" population at all, because each is a different token; they are a separate question and they contribute nothing to the 165. "Closer to Truth" 258 · "Radical Breakthroughs" 157 · "Kuhn Foundation" 25 · "Kuhn Family Foundation" 3 · "How China's Leaders Think" 29. scripts/ocr_variants.py was run on "Robert Kuhn" (one form, "R0bert Kuhn", 0 documents), on "Robert Lawrence Kuhn" (two forms, both 0) and on "Kuhn" (no forms at all); every damaged form that exists in this release was found by reading and none by generating. ↩ ↩² ↩³ ↩⁴ ↩⁵

6. One-page biography, “Robert Lawrence Kuhn”, standing alone in the release with his photograph and no covering message on an adjacent Bates number. Read at the image at 300 dpi by the writer on 2026-09-12. Two short values on the sheet are covered by black bars, one in the description of the firm he co-owned and one in the list of his degrees; nothing is supplied here in their place. https://epstein-data.com/EFTA00800267 p.1. A second copy of the same biography sits inside the House Oversight production at https://epstein-data.com/HOUSE_OVERSIGHT_023666. Its corpus row is not one page but a merge of a long run, so no page pin taken from the row would bind anything; the sheet behind that row is a different matter, and it was read at the image on 2026-09-14. It is one sheet: the biography stands whole on a single page, headed “Robert Lawrence Kuhn”, with the photograph at the top left. A pin to 023666 p.1 therefore binds precisely. An earlier version of this footnote said of the identifier that “its stored text is not one page but a merge of a long run, so no page pin to it would bind anything”. That is true of the row and false of the sheet, and the two are separated here. Nothing above rests on this copy. ↩

7. Gino Yu to Epstein, Subject “Kuhn Interview… Divine intervention.”, “Sent: Wednesday, August 17, 2016 4:49 PM”; the body is one closertotruth.com link and nothing else. Read at the image at 300 dpi by the writer on 2026-09-12. The sender’s address is covered by a black bar. https://epstein-data.com/EFTA02455066 p.1. ↩ ↩²

8. Epstein office schedule for June 2019, headed “2019 une” in the extraction, a run of 4 sheets. The 2 June reminder is on printed p.1, read in the text layer. The 13 June entries are on printed p.3, read at the image at 300 dpi by the writer on 2026-09-12: the sheet prints “Thurs. June 13, 2019 NY”, “Reminder: Rich Kahn on holiday June 13/14, 17 return on Tues. June 18”, “11:00am Appt w/Victoria Brown (Big Think)” and “4:00pm Appt w/Robert Kuhn”, each of the last four appointment lines followed by a black bar where a contact value would stand; nothing is supplied in their place. The run carries its own Bates numbering, printed p.3 bearing EFTA00285756. https://epstein-data.com/EFTA00285754 p.1; https://epstein-data.com/EFTA00285754 p.3. ↩ ↩²

9. Internal Revenue Service determination letter, District Director, Cincinnati, “Date: APR 2 9 1999”, addressed to “THE KUHN FAMILY FOUNDATION / C/O GENEVA COMPANIES”, recognising exemption under section 501(c)(3) and treatment as a private operating foundation under section 4942(j)(3). Identification-number values on this sheet are not reproduced here, and none is completed from any other source. Text layer only. https://epstein-data.com/EFTA00605144 p.1. The run is 5 sheets; p.4 of the same run is a certificate signed by “Dr. Robert L. Kuhn and Daniella R. Kuhn”. ↩ ↩²

10. Three sheets on which the extraction renders the surname Kahn as Kuhn, each read at the image at 300 dpi by the writer on 2026-09-12: Lesley Groff to Epstein, “Schedule July 11-14, 2017”, “Sent: Tue 7/11/2017 9:15:16 AM”, where the sheet prints “TBD Rich Kahn” under Wednesday 12 July and the text layer reads “TBD Rich Kuhn”, https://epstein-data.com/EFTA02216282 p.1; a JPMorgan wire memorandum dated “November 19, 2008” and faxed 22 January 2009, where the sheet prints “Please call Rich Kahn with the Fed Reference number at” followed by a black bar and the text layer reads “Rich Kuhn”, https://epstein-data.com/EFTA01579234 p.1; and a distribution list of “Date: Tue, 13 Jan 2015 11:50:09 +0000” whose To: line prints “Richard Kahn” and whose text layer reads “Richurd Kuhn”, https://epstein-data.com/EFTA00354583 p.1. The other four were read at the image at 600 dpi on 2026-09-12 and settle the same way. https://epstein-data.com/EFTA00442551 p.1, a distribution list of “Date: Sun, 08 Jan 2017 21:09:06 +0000” headed “JE coming to NY tomorrow!”, prints “Rich Kahn” in its To: list where the text layer reads “Rich Kuhn”. https://epstein-data.com/EFTA01372762 p.1 and https://epstein-data.com/EFTA01388065 p.1 are two Bates of the same Deutsche Bank fax of 20 September 2013, a “CONSENT OF THE BOARD OF DIRECTORS OF SOUTHERN TRUST COMPANY, INC.” adopted “on the 19th day of March 2013”: both print the board as “Jeffrey Epstein / Darren K. Indyke / Richard Kahn” where the text layer reads “:Richard Kuhn”. The fourth is the exception that matters to the desk’s standing rule. https://epstein-data.com/EFTA01587972 p.1 is a second Bates of the same JPMorgan transmission as the clean scan cited above: same stamp “Jan 22 2009 12:16 P.01”, same “Project 03-120”, same amount — but it is a machine-reconstructed copy, and its own page image is typeset output of an OCR pass: it renders “JPM FINANCIAL SPONSORS” as “JHM I-INANUIAL SPONSORS”, “JEFFREY EPSTEIN” as “:JEFFREY E Ef$TE1N”, and prints “Pleas call Rich Kuhn with the Fed Reference” in the image itself, cleanly typeset, with nothing to warn the reader that the surname is wrong. Its own typesetting is corrupt in the same line — Please has lost its final e and number is rendered 111,1111ber — but nothing in Rich Kuhn is, which is exactly why the image does not give the reader the warning the rest of the sheet does. An earlier draft of this footnote quoted that line as “Please call Rich Kuhn”, tidying one of the corruptions the sentence exists to demonstrate; it is printed here as it stands. The clean scan of the identical transmission, cited above, prints “Rich Kahn”. Rendering does not always escape the extraction. On this identifier the test that works is not the image but the twin, and this page records that because the desk’s rule — render before you believe the text layer — is necessary and, here, not sufficient. Both copies of that fax carry values of the class this page does not reproduce, and none is reproduced. ↩ ↩²

11. The decomposition of the 165. Bare "Kuhn" returns 1,279 documents and the two-name union accounts for 1,114, leaving 165. All 165 were opened and the line carrying the string was read in each; the bucket totals in the table above are 102 + 7 + 12 + 1 + 13 + 4 + 15 + 11 = 165. A reconciling total proves the buckets partition; it does not prove that any one of them contains what its name says, so each assignment is a judgement about a document that was read, and the buckets are listed here so a reader can overrule them. The 102 assigned to this subject are Epstein’s own lower-case shorthand and the office’s damaged day-sheets, described at ¹⁷; the 7 assigned to Richard Kahn are at ¹⁰, all seven image-verified. The 12 philosopher documents divide: 6 of them are reached by the phrase "Thomas Kuhn" and 6 are not, carrying instead “T. S. Kuhn”, “Thomas S. Kuhn”, “Kuhn TS”, “Kuhn describes” and “a Kuhn-like paradigm shift”. The phrase "Thomas Kuhn" returns 7 documents in all and every one of the 7 sits inside this 165: those 6, plus the Edison Electric Institute man on his own row below. So the philosopher’s footprint in this release is 12 where the phrase alone shows 6 — larger than the phrase returns, but not for the reason an earlier draft of this page gave, which said that none of the 12 was reachable by the phrase. On a page whose method is claims about what a query reaches, that had to be right. The philosopher’s 12 have zero overlap with this subject. The single Edison Electric Institute document is a White House meeting list naming “Edison Electric Institute President Thomas Kuhn”, https://epstein-data.com/EFTA00150058 p.44. The 13 academic documents are bibliography lines citing J. Kuhn, D. Kuhn, R. M. Kuhn, A. Kuhn and Markus Kuhn. The 4 case-citation documents each carry Litton Indus., Inc. v. Lehman Bros. Kuhn Loeb Inc., 122 F.R.D. 433 (S.D.N.Y. 1988). The 15 documents naming other individuals carry twelve distinct people. The 11 noise documents carry no name at all: a page of prison cell codes, a telephone call-detail legend, an airline record field, a bank statement’s damaged line, and two acknowledgements of the Kulin nation which the extraction renders “Kuhn nation”. Adding the 90 of the 102 that no name form reaches to the 1,129-document all-forms union gives the 1,219 used throughout this page. ↩ ↩² ↩³ ↩⁴

12. The top-level sender measurement, run on 2026-09-12 over the first sheet of each of the 1,219 documents, under the rule stated in the body. Counters at a twelve-non-blank-line window: sender parsed 891, sender value absent/barred/blank 202, no sender label 126; his name in the sender value 295, with his forename at the head of the value in 295 of 295; addressed without authoring 259. At windows of 8, 20 and 30 non-blank lines the sender figure is 294, 295 and 295. Under the looser rule that maps the stacked layout positionally, the same population gives 305 (window 12), and 299, 303 and 302 at windows 8, 20 and 30 over the 1,114-document union; the span across both rules at window 12 is 295–305 and the published figure is 295. The same rule over the base phrase alone gives 224 of 776, and over the two-name union 294 of 1,114. scripts/act_rate.py --selftest printed SELFTEST PASS before any figure from it was taken; act_rate.py '"Robert Kuhn"' "Robert Kuhn" "Kuhn" returns sender-field act 280 of 776 (36%), addressed 152, header-less authored 19, press 0, PARSER REFUSED 29; act_rate.py '"Robert Lawrence Kuhn"' "Robert Lawrence Kuhn" "Kuhn" returns 172 of 498, addressed 214, header-less authored 45, press 0, refused 18; the same logic over the 1,114-document union returns act 378, addressed 299, header-less authored 61, press 0, refused 36. A PARSER REFUSED line is a refusal, not a negative, and none of those documents is counted against him anywhere on this page. Duplicate analysis over the 294 union acts: 200 yield a parsable top-level timestamp; those 200 carry 170 distinct timestamps, with 30 timestamps shared by 60 documents; 94 carry no timestamp this page could parse. Year tally over the 170: 2016 24, 2017 74, 2018 35, 2019 37. ↩ ↩² ↩³ ↩⁴

13. Epstein to Richard Kahn, Subject “Fwd: Closer To Truth - Radical Breakthroughs”, “Date: Tue, 01 Aug 2017 12:02:48 +0000”, attaching “Kuhn_Foundation_CTT_JE_–Radical_Breakthroughs-_Incoming_Wire_Instructions.doc”, forwarding Robert Kuhn’s message of “Tue, Aug 1, 2017 at 12:22 AM”. Read at the image at 300 dpi by the writer on 2026-09-12. Kahn’s and Kuhn’s addresses are covered by black bars, and so are the ABA/Routing and SWIFT values on the wire-instruction sheet printed below; none is reproduced here. https://epstein-data.com/EFTA01039322 p.1. ↩ ↩²

14. Epstein to Richard Kahn, Subject “Re: reminder Kuhn 600k”, “Sent: Friday, June 15, 2018 2:44 PM”, body “send 500k today”, carrying Kahn’s message of “Fri, Jun 15, 2018 at 10:03 AM”. The run is 4 sheets and all four were rendered locally at 600 dpi and read by the second fix pass on 2026-09-12. An earlier draft of this footnote stated what stood on printed p.2 and p.3 while only p.1 had been opened, and that is withdrawn; it was also wrong, because the forward of 1 August 2017 begins on p.1 and not on p.2. What stands where, now that every sheet has been read: on printed p.1, Epstein’s header block and body, Kahn’s four-line reply, and the header of Epstein’s forward of 1 August 2017 with his three-word instruction, which prints 150k from gratitud= on this sheet where the copy cited at ¹³ prints it undamaged; on printed p.2, the forwarded message from Robert Kuhn and Epstein’s own earlier message of 9.52 that morning; on printed p.3, the instruction of 14 June 2018, which prints robert kuhn we ne=d to send 600k from foundation . remind me torow where the copy cited at ²⁷ prints it undamaged; on printed p.4, the confidentiality boilerplate and a message-id trailer, and nothing else. Epstein’s own words on p.1 — the body “send 500k today”, the subject “Re: reminder Kuhn 600k”, the Sent value and To: Richard Kahn — are exact at the image. Kahn’s reply is not undamaged, and an earlier draft of this page said it was. The sheet prints =rom gratitude and pape=work; that draft printed from gratitude and paperwork under a footnote stating that every span quoted above printed undamaged on an image it said had been read. The two damaged words are now printed as they stand, in the body and in the Related section, which a second draft repaired in the Related section only. Epstein’s sentence of 9.52 a.m. exists in three copies in this release and not one of them is undamaged, each being damaged in a different place: this sheet’s p.2 prints we=had wired 150 k previosulsy; https://epstein-data.com/EFTA02580751 p.2 prints we had wired 150 k =reviosulsy; https://epstein-data.com/EFTA02581174 p.1 prints we had wired 150 k prev=osulsy. All three were read at the image at 600 dpi on 2026-09-12. The page quotes the sheet it cites and supplies nothing from the other two; a string assembled from two of them would read as undamaged and would stand on no sheet at all. Kahn’s address is covered by a black bar in the header of his quoted message and a value inside his signature block is barred; no barred value and no contact value from that block is reproduced here. The firm name the block carries is not a contact value and is given in Related. https://epstein-data.com/EFTA02581065 p.1, p.2 and p.3. A further copy of the same chain, whose printed p.1 carries the same quoted block, is at https://epstein-data.com/EFTA02580751 p.1. ↩ ↩² ↩³

15. “Jeffrey Epstein spent over $500,000 to fund never-completed episodes of a PBS science series the year before sex trafficking charges”, CNBC, 12 December 2019, https://www.cnbc.com/2019/12/12/jeffrey-epstein-funded-science-tv-episodes-a-year-before-he-arrest.html. The article reports Gratitude America’s Form 990 as its source for the 2018 payment of $500,000 and the 2017 payment of $150,000, gives Kuhn’s figure of $1.4 million for the whole commitment, and quotes him directly. The URL returns HTTP 403 to this desk’s fetcher; the text was read through a syndicated copy at https://usa.timesofnews.com/breaking-news/political/jeffrey-epstein-spent-over-500000-to-fund-never-completed-episodes-of-a-pbs-science-series-the-year-before-sex-trafficking-charges.html, dated 13 December 2019. This is an outside source, not a document in the release, and is cited as the report of a statement rather than as a primary record. ↩ ↩² ↩³ ↩⁴

16. The 15 documents added by the damaged and alternate name forms, over and above the 1,114-document two-name union, were each opened and each is this subject: https://epstein-data.com/EFTA00317670, https://epstein-data.com/EFTA00596833, https://epstein-data.com/EFTA00605144, https://epstein-data.com/EFTA00680590, https://epstein-data.com/EFTA02040617, https://epstein-data.com/EFTA02041149, https://epstein-data.com/EFTA02204714, https://epstein-data.com/EFTA02206736, https://epstein-data.com/EFTA02210296, https://epstein-data.com/EFTA02214427, https://epstein-data.com/EFTA02214954, https://epstein-data.com/EFTA02215291, https://epstein-data.com/EFTA02215504, https://epstein-data.com/EFTA02266007, https://epstein-data.com/EFTA02662499 — each p.1. Twelve of the fifteen are also in the bare-surname residue described at ¹¹; three are not, because the token that reaches them is not Kuhn at all (Kuhr, runn, rlkuhn). ↩

17. Epstein’s bare-surname shorthand, each read in the text layer and each outside both name phrases: “calls kuhn, alain, wendy, karim and eliexer”, 19 October 2016, https://epstein-data.com/EFTA00814679 p.1 and https://epstein-data.com/EFTA02191611 p.1 · “also kuhn jide , kathy , bannon. reid”, 4 January 2019, https://epstein-data.com/EFTA00486454 p.1 · “sat kuhn. thurs deepak.”, 1 April 2019, https://epstein-data.com/EFTA00492495 p.1 · “tell kuhn 1130”, 6 April 2019, https://epstein-data.com/EFTA00493579 p.1 · “Kuhn june 2”, 20 May 2019, https://epstein-data.com/EFTA00495242 p.1 · “sat -30 min before kuhn”, 29 November 2018, https://epstein-data.com/EFTA01015145 p.1. Two Google Calendar notifications in the same bucket carry the entry text “kuhn sultan bottsetin” for 17 March 2017, https://epstein-data.com/EFTA02338249 p.1 and https://epstein-data.com/EFTA02656969 p.1. ↩ ↩²

18. The press test, established by reading rather than by the tool. act_rate.py classes a document as press only if its regex fires inside the first 1,200 characters of the first sheet, and it makes press and header-less-authored disjoint. On the two-name union it reports press 0 and header-less authored 61. All 61 were opened. Thirty-six are press or broadcast captures — CNN, CNN International, Amanpour, Fareed Zakaria GPS, BBC World News, Bloomberg Television, CNBC and CGTN transcripts of his own appearances, and China Daily, South China Morning Post, Shanghai Daily, People’s Daily and International Herald Tribune columns and profiles under his byline, together with one capture of a CCTV programme page. Two are copies of the biography at ⁶; twelve are Epstein office schedules; one is a contact card; two are Skype call logs; eight carry his own writing, of which two open with Epstein’s own words at the head of the sheet (https://epstein-data.com/HOUSE_OVERSIGHT_025191 p.1 and https://epstein-data.com/HOUSE_OVERSIGHT_029134 p.1, both read at the image on 2026-09-14 and both confirming; the site’s PDF store returns HTTP 403 for HOUSE_OVERSIGHT_* identifiers, which is a limit of that route and not of the sheets) with Kuhn’s message quoted below — they carry his writing and are not authored by him at the head, and none of the eight is added to any figure on this page. On the base phrase the same column is 19 documents: twelve schedules, two biography copies, two Skype logs, one press capture (https://epstein-data.com/HOUSE_OVERSIGHT_023668 p.1, a CNBC Squawk on the Street transcript) and two carrying his writing. Re-running the tool’s own regex over the whole first sheet rather than its 1,200-character window fires on 223 documents of the 1,114 union, and on every one of the 223 the only marker it matches is “All Rights Reserved”: 206 are Epstein’s own e-mail confidentiality footer, which ends “copyright -all rights reserved”, and the other 17 are the Darren K. Indyke PLLC footer. Not one of the 223 is a press document. So widening the window is not a better measure either, and the figures above are counts of documents read. ↩

19. Epstein to Gino Yu, no subject, “Date: Thu, 18 Aug 2016 13:58:06 +0000”, body “can you introduce me to kuhn by skype. I ve invited tagg to palm beach for lunch tomrow”. Text layer only. https://epstein-data.com/EFTA00821182 p.1. ↩

20. Robert Lawrence Kuhn to Epstein and Gino Yu, Subject “Jeffrey - Robert”, “Date: Sun, 21 Aug 2016 02:30:55 +0000”, attaching “Kuhn_-Virtual_Immortality-Skeptic_Magazine-21-2-_2016.pdf”, and carrying beneath it Gino Yu’s message of 20 August 2016 12:59 PM and two earlier exchanges of 4 August 2016. Read at the image at 300 dpi by the writer on 2026-09-12. All three correspondents’ addresses are covered by black bars. https://epstein-data.com/EFTA00820611 p.1. The run is 5 sheets. ↩ ↩²

21. Robert Lawrence Kuhn to Epstein, Subject “Re: Jeffrey - Robert”, “Date: Sun, 21 Aug 2016 15:55:22 +0000”, carrying Epstein’s messages of 21 August and Kuhn’s earlier list of on-camera topics recorded with Seth Lloyd. Text layer only. https://epstein-data.com/EFTA00820667 p.1. The run is 5 sheets. ↩

22. Epstein to Gino Yu, no subject, “Date: Sat, 20 Aug 2016 12:49:31 +0000” (“did not hear back from kuhn. send him, a slyep request.”), https://epstein-data.com/EFTA00820546 p.1; Gino Yu to Epstein, “Re:”, “Date: Mon, 22 Aug 2016 16:18:29 +0000” (“Kuhn responded by mail.”), https://epstein-data.com/EFTA00820704 p.1. Both text layer only. Neither document is reachable by either form of his name; both are in the 102 described at ¹¹. ↩

23. Message from Epstein’s office to “Robert Lawrence Kuhr” — the surname damaged in the extraction — Subject “Jeffrey Epstein”, “Sent: Tue 10/4/2016 2:49:00 PM”, signed “Assistant to Jeffrey Epstein”; the sender field is barred and no name is supplied here. This is a block-stacked sheet whose “Sent” field reads “Tue 10/4/2016 2:49:00 PM” with no zone, and no clock reading is given here for the send. Read at the image at 300 dpi by the writer on 2026-09-12. The sender value, the signature name and part of the recipient value are covered by black bars; the visible part of the recipient value prints “Robert Lawrence Kuhn” and the extraction renders it “Robert Lawrence Kuhr”. https://epstein-data.com/EFTA02041149 p.1. ↩

24. Robert Lawrence Kuhn to Epstein, Subject “Shocked in Beijing”, “Sent: Wednesday, November 9, 2016 1:46 PM”, marked “Importance: High”. Read at the image at 300 dpi by the writer on 2026-09-12 and again at 600 dpi on the same date at https://epstein-data.com/EFTA02670720 p.1. His address is covered by a black bar. On the copy read at the image, quoted-printable transport damage replaces a single character at eight points in the body (“=rump”, “=ews”, “=nternationally”, ”=:30”, “=ach”, “=ontributions”, “=o Los Angeles”, “=ext trip”). Every span quoted above is transcribed from that image: the four undamaged ones exactly, and the two damaged ones with their damage shown and glossed. An earlier draft of this page reported that the paragraph beginning “We’re playing in new territory here.” ended on “To be discussed privately.” It does not. On the image that sentence stands mid-paragraph and the paragraph runs on into his travel dates, and the sentence on this page has been rewritten to say so. Both fragments were exact, so no checker could see it. A second copy at https://epstein-data.com/HOUSE_OVERSIGHT_033187 p.1 carries the same body without that damage in its text layer; it could not be rendered, the site’s PDF store returning HTTP 403 for HOUSE_OVERSIGHT_* identifiers, and nothing above is quoted from it. ↩

25. Robert Lawrence Kuhn’s answer of “Wed, Feb 15, 2017 at 1:27 AM” and Epstein’s reply of “Feb 15, 2017, at 4:17 PM” (“is it a charity that received the funding?”), quoted together at the head of https://epstein-data.com/HOUSE_OVERSIGHT_025111 p.1 and https://epstein-data.com/HOUSE_OVERSIGHT_025191 p.1; the answer standing alone at https://epstein-data.com/HOUSE_OVERSIGHT_025106 p.1. All three were read at the image on 2026-09-14 and all three confirm. Both copies of Epstein’s reply print a doubled space between “charity” and “that”, measured at 8× native; the quotation above now preserves it and an earlier version collapsed it. HOUSE_OVERSIGHT_025111 names the attachment “Kuhn Foundation IRS Tax Exempt Determination Letter 1999, 2015.pdf” at the head of the sheet, and 025191 does not; 025106 carries Kuhn’s answer standing alone with no attribution line above it. An earlier version of this footnote said “Text layer only … so none of the three could be rendered.” The site’s PDF store does return HTTP 403 for HOUSE_OVERSIGHT_* identifiers — that is a fact about the route — and all three sheets are ordinary page images that open and read. ↩

26. Robert Kuhn to Richard Kahn, copy to Epstein, Subject “Re: Closer To Truth - Radical Breakthroughs”, “Date: Tue, 01 Aug 2017 20:11:46 +0000”, attaching the two IRS determination letters and carrying beneath it Kahn’s message of “Tue, Aug 1, 2017 at 12:43 PM” and his screen capture of the Internal Revenue Service Select Check results for “Kuhn”, which lists sixteen organisations. Text layer only. https://epstein-data.com/EFTA00666589 p.1; the sixteen-name list at p.2. The run is 6 sheets. A second copy of the same forward is at https://epstein-data.com/EFTA01039342 p.1. ↩

27. Epstein to Richard Kahn, no subject, “Date: Thu, 14 Jun 2018 21:05:31 +0000”, body “robert kuhn we need to send 600k from foundation . remind me torow”. Read at the image at 600 dpi by the second fix pass on 2026-09-12, where the sheet prints the line undamaged; the copy quoted inside the chain at ¹⁴ prints it we ne=d. Kahn’s address is covered by a black bar and is not reproduced here. https://epstein-data.com/EFTA01036120 p.1. A second copy is at https://epstein-data.com/EFTA02638265 p.1. ↩ ↩²

28. Epstein’s message of “Dec 28, 2017, at 7:40 AM” (“great work , but needs some editing, did not like multi verse and sensience. seems more blahblah”), quoted at the head of https://epstein-data.com/HOUSE_OVERSIGHT_029134 p.1; Robert Kuhn’s answer of “Fri, Dec 29, 2017 at 1:45 AM” (“‘Blahblahs’ forbidden in ‘Radical Breakthroughs’. So we need two replacements (for Multiverse as Science and Rare Sentience).”) at the head of https://epstein-data.com/HOUSE_OVERSIGHT_029133 p.1. Both sheets were read at the image on 2026-09-14, at 8× and 10× native respectively, and the wording of both is exact. Two character-level corrections were made against them. First, 029133 opens Blahblahs with a curly left single quotation mark and closes it with a curly right one, and puts a straight apostrophe before Radical and a curly right one after Breakthroughs. An earlier version of this footnote printed a backtick as the opening mark; the backtick is the extraction’s rendering of the curly quote and is not on the sheet. Second, 029134 prints doubled and tripled spaces at four points — after “work”, after the comma that follows it, after “editing,” and after “sensience.” — which an earlier version normalised away and the quotation above now preserves. The site’s PDF store returns HTTP 403 for HOUSE_OVERSIGHT_* identifiers; that is a fact about the route and not about these sheets. ↩

29. Robert Kuhn to Epstein, Subject “Kuhn - Breakthroughs (and Trump)”, “Date: Fri, 29 Dec 2017 20:21:43 +0000”, carrying the revised episode list, the replication-crisis entry and the line about needing more mathematicians. Text layer only. https://epstein-data.com/EFTA00935399 p.1. The run is 4 sheets. ↩

30. Robert Lawrence Kuhn to Epstein, Subject “Math of Fake News”, “Date: Sat, 15 Jul 2017 00:47:21 +0000”, attaching “How_Fake_News_Goes_Viral—Here’s_the_Math_-_Scientific_American.pdf”. Text layer only. https://epstein-data.com/EFTA00697637 p.1. ↩

31. Robert Lawrence Kuhn to Epstein, Subject “Music/Brain”, “Date: Mon, 24 Jul 2017 13:32:22 +0000”, attaching “How_Music_Meets_Mind_-_Nature.pdf”. Text layer only. https://epstein-data.com/EFTA01039503 p.1. ↩

32. Robert Kuhn to Epstein, Subject “Kuhn in NY”, “Date: Fri, 28 Sep 2018 16:11:08 +0000”, forwarding his own message of “September 28, 2018 at 12:05:43 PM EDT” addressed to Epstein and to “Mark Tramo MD PhD”, and carrying beneath it Epstein’s message of “Sep 28, 2018, at 7:28 AM” and Tramo’s of “Thu, Sep 27, 2018 at 3:21 PM”. Read at the image at 300 dpi by the writer on 2026-09-12. Kuhn’s and Tramo’s addresses are covered by black bars. https://epstein-data.com/EFTA01021609 p.1. The run is 2 sheets. ↩

33. Robert Kuhn to Epstein, copy to Peter Getzels, Subject “Kuhn - Breakthroughs - Closer To Truth”, “Date: Wed, 28 Nov 2018 20:16:38 +0000”, attaching “Breakthroughs_-Contributors&Logistics-_11.28.18.pdf”. Text layer only. https://epstein-data.com/EFTA01015208 p.1. Epstein’s forward of it the same afternoon is at https://epstein-data.com/EFTA01015207 p.1. ↩

34. Robert Kuhn to Misha Gromov, copies to Epstein and Peter Getzels, Subject “Closer To Truth - PBS TV - interview - Filming schedule”, “Date: Fri, 15 Feb 2019 01:43:47 +0000”, carrying Gromov’s reply of “Thu, Nov 29, 2018 at 9:49 AM” and Kuhn’s request of “Mon, 26 Nov 2018 10:55:47 -0500”. Read at the image at 600 dpi on 2026-09-12: the quoted paragraph is exact and contiguous on the sheet and carries no transport damage. All three correspondents’ addresses are covered by black bars except Epstein’s own, which is not reproduced here. https://epstein-data.com/EFTA01029442 p.1. The run is 4 sheets. ↩

35. Robert Kuhn to Epstein, Subject “Chaitin”, “Date: Fri, 05 Apr 2019 18:49:23 +0000”, with ten attachments named in the header field. Read at the image at 300 dpi by the writer on 2026-09-12. His address is covered by a black bar. https://epstein-data.com/EFTA01033572 p.1. ↩

36. Lesley Groff to Epstein, Subject “Lunch tomorrow w/Kuhn, Chaitin?”, “Sent: Friday, April 5, 2019 3:27 PM”, https://epstein-data.com/EFTA02634363 p.1; Epstein’s reply, “Re: Lunch tomorrow w/Kuhn, Chaitin?”, “Sent Fri 4/5/2019 3:48:52 PM”, body “yes”, https://epstein-data.com/EFTA02280316 p.1. Both text layer only. Neither document is reachable by either form of his name. ↩

37. Robert Kuhn to Epstein, Subject “Kuhn updates”, “Date: Thu, 31 Jan 2019 02:00:08 +0000”, carrying his own message of “Mon, Jan 21, 2019 at 6:46 PM”, Epstein’s reply of “Jan 21, 2019, at 6:51 PM” and his acknowledgement of “Jan 21, 2019, at 7:37 PM”. Read at the image at 300 dpi by the writer on 2026-09-12 and again at 600 dpi on the same date: the 31 January paragraph quoted above is exact and contiguous on the sheet, which carries no transport damage. His address is covered by a black bar at each of the four points it appears. https://epstein-data.com/EFTA01028406 p.1. The run is 3 sheets. A copy of the 21 January exchange alone is at https://epstein-data.com/EFTA01027542 p.1. ↩ ↩²

38. Jeffrey Epstein to Nicole Junkermann, Subject “do not pass on”, quoted inside her reply of “Sent: Tuesday, January 22, 2019 9:45 AM”; his message is given on the sheet as “On 22 Jan 2019, at 00:59” and its whole body is the sentence quoted above. Read at the image at 600 dpi on 2026-09-12; the sheet prints wider word spacing than is reproduced here and no other departure. Her address is covered by a black bar. https://epstein-data.com/EFTA02626801 p.1. This document does not name Kuhn and is not in this subject’s population; it is cited here only for the sequence, and it is cited on Nicole Junkermann’s page for its own sake. Kuhn’s message of the previous evening is at ³⁷; the copy of that exchange at https://epstein-data.com/EFTA01027542 p.1 was read at the image at 600 dpi on 2026-09-12 and prints the assessment exactly as quoted above, under “On Mon, Jan 21, 2019 at 6:46 PM Robert Kuhn”, with Epstein’s “Neither,?lets wait until something solid” of “Jan 21, 2019, at 6:51 PM” above it and Kuhn’s “OK; let’s see.” at the head of a sheet dated “Tue, 22 Jan 2019 00:37:34 +0000”. The quoted headers on these sheets run five hours behind the +0000 document headers, so 6:46 PM on 21 January and 00:59 on 22 January are a little over an hour apart on one clock and the order of the two messages is not in question on either. ↩

39. Epstein’s message of “9 Jan 2019, at 17:13” (“reviewd today with kuhn your situation”), quoted inside a thread whose top-level message is Nicole Junkermann’s of “Wednesday, January 9, 2019 6:01 PM”, https://epstein-data.com/EFTA02605926 p.1; Epstein to Nicole Junkermann, “Date: Mon, 04 Mar 2019 15:03:44 +0000” (“making some progres in china. i would like you to talk directly to kuhn pick a time.”), https://epstein-data.com/EFTA01031706 p.1; Epstein to Nicole Junkermann, “Re:”, “Date: Fri, 08 Mar 2019 17:20:59 +0000” (“lets me confirm does sunday work as welll kuhn is in china”), https://epstein-data.com/EFTA01030895 p.1; Epstein to Darren Indyke, “Sent Sun 3/10/2019 2:22:41 PM” (“conference call in for kuhn ?”), https://epstein-data.com/EFTA02278852 p.1. All four text layer only, and none of the four is reachable by either form of his name. ↩

40. Epstein to Nicole Junkermann, no subject, “Date: Fri, 04 Jan 2019 01:55:53 +0000”, the quoted words standing in a longer paragraph of advice on the arbitration. Text layer only. https://epstein-data.com/EFTA01009490 p.1. The same sentence is carried on https://epstein-data.com/EFTA01009258 p.1, https://epstein-data.com/EFTA01009260 p.1, https://epstein-data.com/EFTA02606254 p.1 and https://epstein-data.com/EFTA02607494 p.1. This message is also quoted, for a different purpose, on Nicole Junkermann’s page. ↩

41. Bebe Avdiu to a barred recipient, copy to Lesley Groff, Subject “Jeffrey Epstein”, “Date: Mon, 10 Apr 2017 14:01:15 +0000”, signature block “Bebe Avdiu / Legal Assistant / DARREN K. INDYKE PLLC”. Text layer only. The recipient value is barred and the telephone line in the signature block is covered on the sheet; neither is supplied here. https://epstein-data.com/EFTA00449322 p.1. ↩

42. Message from Epstein’s office to “Robert Lawrence Kuhn”, Subject “Jeffrey Epstein”, signed “Assistant to Jeffrey Epstein” under a barred line. Read at the image at 600 dpi on 2026-09-12. The sheet prints Sent: Wed 10/5/2016 1:25:07 PM and the recipient as Robert Lawrence Kuhn; the text layer of the same sheet damages the Sent value to “ed 1 / / 16 1:25:07 PM” and the recipient to “ce Kuhn”, and an earlier draft of this page reported from that text layer that no date could be read. The image carries the date. The sender field is covered by a black bar and nothing is supplied in its place. The body prints at his home followed by an ellipsis and then Woody Allen; the text layer renders that ellipsis as an em dash. The em dash is not on the document, and an earlier draft of this page printed it. https://epstein-data.com/EFTA02040496 p.1. A second, independent copy of the same message, quoted inside a forward to Epstein, is at https://epstein-data.com/EFTA00317567 p.1 and was also read at the image at 600 dpi: it prints the ellipsis likewise, gives the message as sent “On Oct 5, 2016, at 9:25 AM”, and carries Kuhn’s reply of “October 5, 2016 at 9:54:31 AM EDT” asking whether he might bring his wife, who is not named here. The two clocks differ by five hours, the quoted headers on these sheets running behind the +0000 document headers; the two sheets are the same message and not two. Kuhn’s address is barred on both sheets; no address from either sheet is reproduced here. ↩

43. Merwin Dela Cruz to Epstein, Subject “Sir, Mr Kuhn is here”, “Sent: Thursday, June 15, 2017 12:56 PM”, body “Thank you / Merwin Dela Cruz”. Text layer only. https://epstein-data.com/EFTA02646771 p.1. ↩

44. Epstein office schedules naming him. The three EFTA sheets were rendered locally at 600 dpi and read by the second fix pass on 2026-09-12, and on all three the image prints the forename correctly where the extraction damages it. On https://epstein-data.com/EFTA02206736 p.1 the sheet prints “4:00pm Appt w/Robert Kuhn” under “Mon. Feb. 13, 2017 NY”, followed by a black bar; the extraction renders it “4:00pm Appt w/Roberi Kuhn.” and the full stop is the extraction’s, not the sheet’s. On https://epstein-data.com/EFTA02210296 p.1 the sheet prints “11:00am Robert Kuhn” under “Thurs. April 13, 2017 NY”; the extraction renders it “11:00am Robed Kuhn”. On https://epstein-data.com/EFTA02214427 p.1 the sheet prints “9:00am BREAKFAST w/Robert Kuhn” under “Thurs. June 15, 2017 NY”, followed by a black bar; the extraction renders it “9:00am BREAKFAST w/Roberi Kuhn”. The final glyph of the forename was measured on each of the three rather than eyeballed: at 600 dpi it is continuous ink from above the x-height down to the baseline, with a crossbar at the x-height and no detached dot above, which is a t and is neither an i nor a d. This is the desk’s standing trap running the other way — a match against the extraction is no proof, and neither is a mismatch. The other three sheets were read at the image on 2026-09-14, and the forename prints correctly on each of them too: “TBD Robert Kuhn?”, standing at the head of https://epstein-data.com/HOUSE_OVERSIGHT_002625 p.1 and again under “Fri. Aug. 4, 2017 NY-Bedford-PB” on https://epstein-data.com/HOUSE_OVERSIGHT_002632 p.1; “Reminder: Robert Kuhn is hoping to see you Friday if you will be in NY”, followed by a black bar, under “Thurs. Aug. 3, 2017 NY” on https://epstein-data.com/HOUSE_OVERSIGHT_002632 p.1; and “Reminder: Robert Kuhn to depart NY around 10th”, August 2017, at the head of https://epstein-data.com/HOUSE_OVERSIGHT_002622 p.1. So the claim that the forename prints correctly now rests on six image readings rather than three images and three text layers. Three of these six carry a damaged forename in the extraction — the three EFTA sheets, which are for that reason reachable by neither form of his name. An earlier draft of this footnote said five of six; that was wrong and is corrected here. An earlier version of this footnote said the HOUSE_OVERSIGHT_* sheets “could not be rendered”. The /pdf/ route on this site does return HTTP 403 for that prefix; the sheets are ordinary page images and read. ↩

45. Robert Kuhn to Lesley Groff, Subject “Re: Jeffrey Epstein”, “Sent Thur 5/30/2019 1:34:52 AM” (“Greetings from Beijing!”), https://epstein-data.com/EFTA02287527 p.1; his reply of “Sent: Thur 5/30/2019 11:09:03 AM”, https://epstein-data.com/EFTA02289016 p.1; and the closing message, “Sent Thur 5/30/2019 2:17:07 PM”, body “Great; 4:00 pm, June 13. Looking forward / Robert”, https://epstein-data.com/EFTA02287474 p.1. All three are block-stacked sheets whose time fields carry no zone; the quoted messages inside them are rendered in more than one local zone and the order is taken from the nesting. Text layer only. ↩

46. "Closer to Truth" is the name of a television programme and is not a measure of this man. The phrase returns 258 documents, of which 232 are inside the two-name union and 26 are outside it. All 26 were opened: 7 are copies of a different person’s curriculum vitae listing “PBS-TV Closer To Truth” among his media appearances; 1 is Deepak Chopra writing “My review of Krauss book Are we closer to Truth ?”; 1 is another correspondent’s wordplay, “Lawrence is closer to truth than he realizes”; and the remaining 17 are documents of this project that the name phrases cannot reach — his own messages with the sender value barred or damaged, and Epstein’s forwards of the project’s budget and pre-production attachments. An appearance on the programme, a mailing-list copy, a forwarded episode link and this man acting are four different things, and the figure 258 mixes them. "Radical Breakthroughs" returns 157 documents, 145 of them inside the union. Neither search returns any accounting for the money, any grant report, or any document dated after 2 June 2019. ↩