I found myself seated at the counsel table alongside Roman Sterlingov, a 35-year-old man, during the biggest trial ever conducted on Bitcoin-mixer money laundering. Although we referred to him as “Mr. Sterlingov” throughout the trial, I personally knew him as Roman. He was the defendant, and we anxiously awaited the jury’s verdict.
On March 12, the verdict was delivered: “guilty.” As each of the four counts of the indictment were read, the words resonated within me like a physical blow to the gut. To process this devastating news, my mind instinctively shifted towards formulating strategies for an appeal.
The trial spanned over four weeks, during which I dedicated an entire day to testifying and spent a year working on the case. The focal point of the trial was Bitcoin Fog, the largest mixer in Bitcoin’s history.
Bitcoin Fog had facilitated the processing of approximately 1.2 million Bitcoin throughout its existence. Allegedly, hundreds of millions of dollars in drug money from darknet platforms such as Silk Road and AlphaBay had been laundered through this platform. The prosecution accused Roman not only of using Bitcoin Fog but also of running it.
In their defense, Tor Ekeland and Mike Hassard, Roman’s legal counsel, fought tenaciously, reminiscent of Paul Newman’s character in “The Verdict,” but with a technological twist.
The prosecution’s case closely mirrored the original indictment. It revolved around a Bitcoin transaction from Sterlingov’s Mt. Gox account that eventually ended up in a Bitcoin wallet. The identity of the wallet owner and the holder of its private key remained unknown. Subsequently, a series of transactions were traced back to the purchase of a Bitcoin Fog clearnet site, which provided instructions on accessing Bitcoin Fog on the darknet.
It was posited that Sterlingov may have sold Bitcoin to an individual who acquired the Bitcoin Fog website, or perhaps that individual later sold Bitcoin to another person, leading to a chain of transactions that ultimately culminated in the purchase of the domain.
The prosecution heavily emphasized Sterlingov’s use of Bitcoin Fog. He admitted to utilizing the platform for privacy purposes, but the government’s claim that he had sent 2,700 Bitcoin through Bitcoin Fog was disputed during my testimony. I demonstrated that the true operator of Bitcoin Fog could have earned between 24,000 and 36,000 Bitcoin based on the platform’s fees, amounting to hundreds of millions of dollars. This figure aligned with the testimony of a government witness, Larry Harmon, who had operated a related Bitcoin mixer called Helix. However, the government’s IRS witness revealed that Roman never spent more than $60,000 annually, resided in a one-bedroom apartment, and was never worth more than $1.8 million during the ten-year surveillance period.
The “Perry Mason” moment arrived when the defense expert, Jeff Fishbach, uncovered a crucial flaw in the government’s evidence. A screenshot of a text message chain, initially presented as proof of a money laundering plan by the defendant, turned out to be a mere image from an e-book that Roman had been reading on his computer. The prosecution apologized for this error during closing statements, assuring the court that it was an isolated incident.
Interestingly, C. Alden Pelker and Chris Brown, the prosecutors in this case, had previously advised against basing cases solely on tracing evidence in a publication for the Justice Department. They recommended the use of corroborating evidence, such as the possession of a private key to Bitcoin addresses containing illicit funds. This advice was logical, given that academic literature suggests that Chainalysis heuristics, which were relied upon in this case, are incorrect 90% of the time. Relying on such flawed evidence when someone’s freedom is at stake is a grave mistake, and yet the prosecutors in this trial committed the very error they had cautioned against.
The Chainalysis tracing in this case relied heavily on two tracing heuristics: the “co-spend” heuristic, which assumes that Bitcoins spent together originate from the same user, and the “peel chain” heuristic, which assumes that unspent Bitcoins are linked in a chain where the larger transaction represents the spender’s “change.” However, these heuristics are easily defeated, whether by splitting a dinner bill with a friend using Bitcoin or by sending a larger transaction to another person within the chain. Moreover, in the early years of Bitcoin, off-chain transactions involving the transfer of private keys were more common, rendering these heuristics ineffective.
Roman was an early adopter of Bitcoin, which, combined with his use of Bitcoin Fog for privacy, his possession of a Russian passport, and his affinity for computers, made him an easy target for being associated with the operation of Bitcoin Fog. In that regard, he was perhaps the most unfortunate person I had ever encountered.
J.W. Verret is an associate professor at George Mason University’s Antonin Scalia Law School. He is a practicing crypto forensic accountant and also practices securities law at Lawrence Law LLC. He serves as a member of the Financial Accounting Standards Board’s Advisory Council and is a former member of the SEC Investor Advisory Committee. Additionally, he leads the Crypto Freedom Lab, a think tank dedicated to advocating for policy changes that protect freedom and privacy for crypto developers and users.
This article aims to provide general information and should not be construed as legal or investment advice. The opinions expressed in this article are solely those of the author and do not necessarily reflect the views and opinions of Cointelegraph.