Hong Kong law now accepts that crypto-assets are capable of being property and can, in principle, be held on trust. That matters. But in most disputes, it is no longer the point that decides the case.
The real issues usually arise later: who controlled the wallet, whether the platform relationship was trust-based or purely contractual, whether the assets remained identifiable after pooling or onward transfers, what evidence is available, and what relief the court can grant that will actually work in practice.
That is where Hong Kong crypto disputes are now likely to be fought. Recognition matters, but it is only the starting point.
That was an important step. Without it, many fraud and insolvency claims would risk being reduced to personal claims only. But the practical significance of Gatecoin is narrower than it is sometimes presented. It does not answer whether a particular customer relationship was trust-based, whether a claimant can trace into mixed or substituted assets, or whether relief against an exchange will be workable.
So the threshold issue is now largely behind us. The harder questions are factual, evidential and structural.
Start with custody
In practice, the first question is usually simple: how were the assets actually held?
That requires both a legal and an operational analysis. The terms may describe “custody” or “client assets”, but the court will also look at what the platform was entitled to do, how the assets were recorded, and whether they were in fact segregated or pooled.
The key questions usually include:
(a)were the assets held in a segregated or omnibus wallet;
(b)did the customer have rights to an identified fund or only an account balance;
(c)who controlled the private keys or signing arrangements;
(d)was the platform free to use the assets for its own purposes; and
(e)did the operational reality match the documents.
Those points often determine whether there is a proprietary claim at all.
Control is not ownership
Parties often focus first on who held the private key. That is understandable, but it is only part of the picture.
Control of a wallet may show who could move the asset. It does not, by itself, determine beneficial ownership. The key-holder may be owner, trustee, custodian, nominee or agent. A wrongdoer who gains access to the wallet acquires control, not better title.
For that reason, practitioners should keep separate: the wallet address, the person with technical control, the platform entity dealing with the customer, and the person claiming beneficial ownership. In urgent applications, that distinction matters.
Trust still has to be proved
A crypto-asset may be capable of being trust property, but that does not mean a trust will readily be found.
That is one of the most useful parts of Gatecoin. Although the Court accepted that cryptoassets are capable of being held on trust, it concluded that the relevant customer relationship was not trust-based under the relevant contractual terms. The documents and overall structure pointed away from fiduciary obligations.
The practical point is straightforward. A claimant cannot simply rely on the fact that the asset is property. A trust case still depends on the contractual terms, the surrounding representations, and the operational arrangements.
In practice, a stronger trust case usually requires some combination of:
(a)language consistent with safekeeping or segregation;
(b)limits on the platform’s ability to use the assets;
(c)a clear custodial purpose;
(d)records and accounting consistent with that structure; and
(e)sufficiently certain subject matter.
If those features are weak, personal claims should usually be pleaded alongside any proprietary case.
JPEX shows the value of a focused case
The decision in Chan Wing Yan v JP-EX Crypto Asset Platform Pty Ltd[2024] HKDC 1628 shows how a practical trust case can succeed.
There, the Hong Kong District Court granted default judgment on an express trust case concerning JPEX. The claimants relied on a focused factual case: the platform controlled the private keys, represented that assets would be safeguarded in cold wallets, held them for a defined custodial purpose, and moved the deposited USDT away almost immediately.
That combination mattered. The case was not driven by abstract submissions about the nature of cryptoassets, but by a clear account of what was promised, how the assets were to be held, and what in fact happened to them.
That is usually the better approach.
Pooling changes the case
Pooling is often where proprietary claims become difficult.
Many platforms sweep customer deposits into omnibus wallets and record entitlements only through internal ledger entries. Once that happens, the claimant must do more than point to the original deposit. The real question becomes whether the claimant can assert rights against an identifiable pooled fund, or only a personal claim against the platform.
That issue should be confronted early. A claimant seeking proprietary relief should be ready to explain:
(a)what happened to the deposited assets after receipt;
(b)whether they were mixed with other customer or house assets;
(c)whether the claim is to the original asset, a pooled fund or a substitute asset; and
(d)how the alleged proprietary interest survived the mixing.
If those points are left unclear, urgent relief becomes much harder to justify.
Exchanges are not automatic trustees
Where assets pass through an exchange, the exchange is often the obvious target. It may be the only party able to identify the account holder, preserve records or halt further movement.
But receipt by an exchange does not automatically make the exchange a trustee, nor does it mean traceable proceeds can easily be identified within the exchange’s systems.
That is the warning from Piroozzadeh v Persons Unknown[2023] EWHC 1024 (Ch), where an interim proprietary injunction was discharged because the application had not fairly addressed the exchange’s pooling model, possible title defences, and the practical difficulty of identifying specific proceeds within a high-volume exchange environment.
The lesson is clear. If relief is sought against an exchange, the claimant should explain why that exchange is a proper target, what asset or information it is said to hold, and whether the relief sought can realistically be implemented.
Tracing evidence must be usable
A tracing report may look impressive and still fall short.
In crypto cases, there is a recurring tendency to treat blockchain analysis as if it were legal tracing. It is not. Transaction mapping may show movement between addresses, but the court still needs to know what legal conclusion the evidence is said to support.
That is why D’Aloia v Persons Unknown[2024] EWHC 2342 (Ch); [2025] 1 WLR 821 is important. Although the Court accepted that USDT was property and could in principle retain its identity through transfers, it criticised aspects of the tracing methodology as unclear, inconsistent and mathematically unsound.
The practical point is that expert evidence must be built for litigation, not just investigation. It should explain clearly:
(a)what asset is said to have been followed or traced;
(b)whether the case is one of following or tracing into substitutes;
(c)how pooling and recombination were addressed;
(d)what assumptions were made; and
(e)what the limits of the analysis are.
If the expert cannot explain the route from blockchain data to legal remedy, the report may carry limited weight.
Build the case around the remedy
Crypto disputes move quickly, but speed is not enough. Evidence, defendants and relief must be aligned from the outset.
In practice, that usually means:
(a)preserving source material immediately, including wallet addresses, transaction hashes, screenshots, exchange communications, terms and bank records;
(b)obtaining early forensic input with a clear legal brief;
(c)identifying defendants by function, rather than treating the fraudster, exchange, custodian and innocent intermediary as interchangeable;
(d)seeking relief that the respondent can actually comply with; and
(e)addressing situs, service and jurisdiction at the start, not later.
In these cases, a technically strong complaint can still fail if the application is not structured in a way the court can act on.
Prevention still matters
For platforms, custodians and institutional users, the most effective disputes strategy still begins before any dispute arises.
The legal documents, technical architecture and internal operations should point in the same direction. If the platform says assets are segregated, the operating model should support that. If the relationship is intended to be debtor-creditor rather than trust-based, the documents and internal treatment should say so clearly.
At a minimum, parties should be clear on whether customer assets are segregated or pooled, whether use is permitted, who controls key material, how entitlements are recorded, and what happens on insolvency or a court order. In this area, inconsistency between paper and practice can become the central problem in the dispute.
Cross-border caution
Cases involving the Mainland require added care. Mainland courts have shown some willingness in civil disputes to recognise that crypto-assets may have the characteristics of property or “virtual property”. But that sits alongside a far more restrictive regulatory position on cryptocurrency trading and related activity.
The practical point is that private-law recognition and regulatory permissibility are not the same thing. In cross-border matters involving Hong Kong and the Mainland, that distinction should be built into the strategy from the beginning.
Conclusion
Hong Kong has answered the threshold question. Crypto-assets are capable of being property and may in principle be held on trust. But in practice, disputes will rarely turn on that point alone.
The real fight is likely to be over custody, control, pooling, tracing, documentation and workable relief. Parties who treat “property” as the end of the analysis may get little value from the recognition. Parties who focus early on how the assets were held, what the documents really say, where the evidence sits and what relief can realistically be enforced will be in a far stronger position.
Recognition opened the door. The hard work starts after that.
Beijing ICP No. 05019364-1 Beijing Public Network Security 110105011258
