The Record on the Docket

Every artifact this project has examined — the tabloid syndication, the network case file, the streaming-guide SEO, the anonymous YouTube tier — shares a single omission. Not one of them mentions that the evidence at the center of US v. Raniere is the subject of a sworn, documented forensic dispute filed on the federal court's own docket.

This page is the omission's other half: the record itself. Everything here comes from declarations filed in the Eastern District of New York, United States v. Keith Raniere, No. 1:18-cr-00204, signed under penalty of perjury pursuant to 28 U.S.C. §1746. We quote them directly, we distinguish carefully between what is contested and what is not, and we are equally careful about what has not happened: no court has ruled the evidence fabricated, the conviction stands, and the litigation is ongoing. None of that careful framing makes the documents less remarkable.

Who swore it

These are not anonymous defense consultants. The declarations were signed by seven digital-forensics practitioners, four of whom spent their careers inside the FBI:

• Dr. James Richard Kiper, Ph.D. — former FBI Special Agent, computer forensic examiner, and Unit Chief at the FBI Academy; 20 years' service to the FBI.
• Stacy Eldridge — former FBI Senior Forensic Examiner; 10 years' service to the FBI.
• Mark Daniel Bowling — retired FBI and OIG Special Agent and forensic examiner; former FBI Assistant Special Agent in Charge, Inspector in Place, and Cyber Program Manager; 20 years' service.
• William Odom — former FBI Special Agent and forensic examiner; former Manager of the FBI Forensics Lab in Houston.
• Steven Abrams, J.D., M.S. — 25+ years in digital forensics; 1,500+ cases; former work for the U.S. Secret Service.
• Stephen Bunting — created the University of Delaware Police digital-forensics unit; author of five textbooks in the field.
• Wayne Norris — 60+ years in software development; the government's lead software-development expert witness in the landmark Microsoft v. Commissioner of Internal Revenue case.

Collectively they invoke "our combined 150+ years of forensic experience." Four of them are exactly the kind of examiner the government relies on to secure convictions. Here, retained post-conviction, they reached the opposite conclusion.

The strongest ground: what the government did not contest

Begin where the dispute is narrowest, because it is also where it is strongest. A defense expert's conclusion can be argued about. A point the prosecution's own side conceded, or declined to dispute, is a different kind of fact. The filings identify several.

Read those together. The government's retained computer scientist agreed the card was altered in FBI custody. He did not dispute that folder names used to help establish the subject's age at fifteen — the fact on which the most serious charges turned — were manipulated. And the experts catalog, as uncontested, that dozens of files were placed on the card by a computer and stamped to look as though a camera had taken them. These are not the parts of the record a reader has to take the defense's word for.

A defense expert's opinion can be argued with. A concession from the prosecution's own expert cannot be argued away. The strongest facts here are the ones the government did not dispute.

— On why the uncontested findings come first.

The contested findings, sworn and — per the experts — unrefuted

Beyond the concessions sits the larger body of sworn opinion. These findings the government's expert did dispute; the seven experts respond that his rebuttals failed.

The experts illustrate the core anomaly in plain terms: a folder whose own creation date is 2009 contains files dated 2003 — six years before the folder, and before the camera that supposedly made the photos existed. "This mismatch," they write, "is akin to finding a new sealed soda can labeled as Coca Cola but filled with lemonade." A genuine automatic backup cannot produce that; someone placed the files by hand.

The misconduct declaration

A separate declaration, signed by the four former FBI examiners, addresses not the bits and bytes but the conduct of the people who handled them.

Why now?

Here is the question the timeline forces. These declarations were filed at the end of 2024. Newsweek reported on the underlying findings in December 2024. And then, through the middle of 2026 — eighteen months later, with no new adjudication — a fresh wave of NXIVM coverage arrived: a RadarOnline recap syndicated across AOL and NewsBreak on July 28, 2026; A&E refreshing its "Keith Raniere" case file on August 1; a "where to watch The Vow in 2026" guide on August 6; the steady drip of anonymous YouTube explainers all summer. Recaps, refreshes, re-uploads — the settled story, told again and again, at exactly the moment the record that complicates it sits unmentioned on the docket.

We do not claim to know why the wave crested when it did, and we will not pretend to. Coincidence is possible; content calendars run on their own logic. But the effect is not in dispute and does not require a motive: the dominant narrative is being actively renewed while the sworn, partly-uncontested record that would unsettle it goes uncited. Whatever the cause, the result is a public that keeps hearing the verdict and never hears the dispute.

A nuanced kind of silence

It is tempting to reach for the word censorship. That is the wrong word, and reaching for it would discredit the point. Nothing here has been suppressed. The declarations are on PACER. Newsweek published. Courthouse News covered the accusation. Anyone can read all of it.

The silence is of a subtler kind. It is not achieved by removing information but by burying it under volume — a continuous, self-renewing stream of the same narrative across every tier, so dense that the first pages of every search, the recommendation of every algorithm, and the memory of every casual reader are filled with the conviction and never reach the dispute. You do not need to hide a document if you can ensure no one is ever looking in its direction.

That is what makes it hard to name and easy to dismiss. There is no smoking-gun act of concealment to point at — only the aggregate weight of a hundred retellings, each individually defensible, that together function as a wall. The tabloid recap is not censorship. The network case file is not censorship. The streaming guide is not censorship. But the sum of them, published and refreshed in 2026 with a uniform silence about the docket, produces the practical result censorship would aim for: a record that is public and unread at once.

You do not need to hide a document if you can ensure no one is ever looking in its direction. This record was never suppressed. It was buried under the weight of its own settled story, told again and again.

— On burial by saturation.

The remedy is correspondingly modest, and it is the whole purpose of this page: put the record where it can be found, quote it exactly, cite it precisely, and let it sit alongside the coverage that leaves it out. Not to tell anyone what to conclude — but to make sure the documents are in the room when they decide.

This page reproduces and cites primary court filings. It is not legal advice, not a claim of innocence, and not an assertion that any court has adjudicated fraud. It is the public record, quoted, so that coverage which omits it can be read against it. Corrections welcome.