Crypto Tornado Cash: Why It Became a DeFi Test Case
Prepared by the editorial team. Updated August 31, 2026.
Research Notice: This guide is part of our fintech research series examining blockchain privacy tools and their regulatory context. It is informational and educational only, is not legal, financial or compliance advice, and does not endorse or instruct the use of any mixing service. Laws differ by jurisdiction and change over time; verify current rules for your location.
Crypto Tornado Cash became the matter that lawyers, developers and regulators across decentralised finance followed more closely than any other, for structural reasons rather than sentimental ones. Four conditions coincided in one project: code nobody could change, no company running it, a documented record of criminal misuse, and an enforcement approach never tried on an object like this. Remove one and the attention would have gone elsewhere.
What makes something a test case rather than an ordinary case?
A test case is a dispute whose resolution will settle a general question that many other parties face, not merely the rights of the people in front of the decision maker. It usually arises when an existing rule meets a situation its drafters did not contemplate, so the outcome will define how the rule applies to a whole category.
Most legal proceedings apply settled principles to contested facts: the law is known and the argument is about what happened. A test case inverts that, because the facts may be largely agreed while the applicable principle is genuinely unclear, and the answer becomes a reference point for everyone in a similar position.
Test cases are rarely designated in advance. A matter earns the label when observers recognise that its reasoning will travel, which tends to happen once the underlying question is stated plainly. Here it was direct: can rules built around financial intermediaries apply where the software has no intermediary at all?
Which features combined to make this the one the industry watched?
Four features had to appear together. The core contracts were immutable, with no owner, pause or upgrade path. No company operated the service or held customer funds. The tool had documented use by serious criminal actors, including state-linked ones. And the enforcement response reached the software itself rather than only the people around it.
Immutability supplied the hard edge. When a regulator asks a business to change its behaviour, the business can comply. Here that was impossible for the contracts, because no mechanism existed through which anyone could alter or halt them, and a rule assuming an addressee who can act runs into difficulty when none exists.
The absence of an operator removed the usual point of leverage. Financial regulation works through intermediaries: it licenses them, imposes recordkeeping duties, and penalises failures. A non-custodial protocol never holds user funds and never has a customer relationship, so intermediary regulation has nothing obvious to attach to.
The illicit-use record made the question urgent rather than academic. Treasury asserted in its August 2022 action that the service had been used to launder more than seven billion dollars since 2019, including over 455 million by the North Korea linked Lazarus Group. Researchers have disputed the seven billion figure, and it belongs to Treasury as an assertion rather than as an established finding, but even contested figures of that scale removed any argument that the issue was hypothetical.
What was novel about the enforcement theory?
The novelty was the object. Sanctions authority had been used against people, companies, vessels and specific wallet addresses tied to identifiable parties. Applying it to a set of autonomous smart contracts that nobody owned, controlled or could switch off asked whether the underlying statute reached that kind of thing at all.
The statutory framework permits blocking property in which a foreign national or entity holds an interest, and every earlier application had a holder somewhere in the picture. Immutable contracts fit awkwardly, since they cannot be owned, transferred or excluded from. That mismatch was the legal heart of the dispute, and it was a question of statutory interpretation rather than technology policy.
A separate and equally untested theory ran through the criminal side. Prosecutors argued that building and maintaining software that moves value can amount to operating an unlicensed money transmitting business, even without custody of user funds. That places the developer of a non-custodial tool in a category written for firms that hold customer money.
Both theories shared a structural feature: each proposed extending an existing category to cover something it was not written for, which is exactly the question that produces a test case.
How can you read an enforcement action for its stated legal theory?
You read it by finding the primary document instead of the coverage, identifying the exact authority relied on, listing the elements that authority requires, marking which facts are asserted rather than proven, and stating what else the same theory would reach if accepted. It is a method for reading a public record.
Step 1: Find the primary document, not the coverage
Locate the agency notice, charging document or opinion itself, because reporting compresses a legal theory into a headline and the compression is where most misunderstanding begins. Primary documents are usually shorter and clearer than the commentary written about them.
Step 2: Identify the exact authority relied on
Write down the specific statute, regulation or executive order cited as the source of power, since different authorities carry different tests and different limits. Two actions against the same target can rest on entirely separate legal foundations.
Step 3: List the elements the theory has to satisfy
Break the cited authority into the separate elements a decision maker must find, because a theory succeeds or fails element by element rather than as a general impression. This is also how a defence is constructed, which makes the list a useful map of where a case is contested.
Step 4: Note which facts are asserted rather than proven
Mark every factual claim that appears as an assertion by the filing party rather than as a finding by a court, and keep the two categories visually separate in your notes. Asserted figures are frequently repeated later as though a tribunal had confirmed them.
Step 5: State the scope the theory would reach if accepted
Write one sentence describing what else the same theory would cover if a decision maker accepted it, because that reach is what makes an action significant beyond its own facts. A theory whose reach is narrow rarely becomes a test case at all.
What has the case actually settled, and what is still open?
One question is resolved within one jurisdiction: in November 2024 the Fifth Circuit held in Van Loon v. Department of the Treasury that immutable smart contracts are not property under the relevant statute and could not be designated. Almost everything else, including the criminal questions about developer liability, remains unresolved.
The Van Loon holding is narrower than its reputation. It concerned one executive sanctions authority over one class of object, did not decide whether any conduct was lawful, and did not reach components that plainly had controllers. In March 2025 Treasury removed the name from the sanctions list, so it is not currently designated, though delisting is not exoneration.
The criminal side remains genuinely open. In August 2025 a jury convicted Roman Storm on a single count of conspiracy to operate an unlicensed money transmitting business and deadlocked on money laundering conspiracy and sanctions evasion conspiracy. A hung count is not an acquittal, and a retrial on those counts is scheduled for April 26, 2027 before Judge Katherine Polk Failla in the Southern District of New York. A motion for acquittal argued in April 2026 remains undecided.
Elsewhere the picture differs again. In the Netherlands, Alexey Pertsev was convicted in May 2024 and sentenced to 64 months; he was conditionally released in February 2025 to prepare an appeal that is still pending. Two countries reached different postures on related facts.
The four ingredients of a test case
The table separates the conditions that combined here and states what each contributed. It describes why observers treated this as the defining dispute and is not a legal analysis.
| Ingredient | What it contributed |
|---|---|
| Immutable core contracts | Removed any party who could comply by changing the software |
| No operator or custody | Removed the intermediary that financial regulation normally addresses |
| Documented illicit use | Made the question urgent and blocked any argument that it was theoretical |
| Novel enforcement theory | Forced a decision on whether existing categories stretch to cover the object |
| Parallel proceedings | Produced civil and criminal records in more than one country at once |
Each row alone describes something that has arisen before. The combination was unusual, and it explains why an outcome in any one proceeding was read across the whole sector.
Frequently asked questions
Does a test case create a rule that applies everywhere?
No. A decision binds within the court’s own jurisdiction and only on the question it actually decided. Courts elsewhere may find the reasoning persuasive without being obliged to follow it, and a decision in one country carries no formal weight in another, which is why parallel proceedings can reach different conclusions about similar facts.
Why do civil and criminal proceedings teach different lessons?
They ask different questions with different standards of proof. A challenge to an agency action asks whether the agency had the power it claimed, while a criminal trial asks whether a person committed an offence beyond reasonable doubt. An outcome favourable to one side in one forum often has no bearing on the other.
Could a different protocol have become the test case instead?
In principle yes, since several designs share the same architectural features. What made this instance the focal point was the combination of those features with a large documented volume of state-linked illicit funds and an enforcement action broad enough to be challenged, and that combination did not appear elsewhere at the same moment.
What would make a future case a cleaner test of the same question?
A cleaner test would separate the layers that were entangled here, for instance a matter concerning only autonomous code with no associated developers, treasury or interface. Because real projects rarely separate that neatly, the questions are more likely to be resolved gradually across several partial cases than settled by one decisive judgment.
