1. A custodian's case for off exchange settlement
Darren Jordan, chief commercial officer of digital asset custodian Komainu, argued in a column published on November 18, 2026, that institutions should stop leaving client assets with the venues where they trade. He pointed to October 29, when about $6.3 billion left Binance's hot wallets in 20 minutes, and presented off exchange settlement as the remedy.
The piece is an opinion, and it comes from an executive whose company sells custody services, so its conclusions align with his firm's business. With that context stated, the argument he sets out is a structural one rather than a criticism of any single exchange. His claim is that the way crypto markets grew up left trading and safekeeping fused together, and that the arrival of large regulated investors makes that arrangement increasingly difficult to defend.
Jordan's central image is borrowed from traditional finance. In established markets, he writes, "Nobody expects the trading venue to double as the vault." The rest of his column explains why he believes digital asset markets should adopt the same expectation, how that can be done without giving up access to exchange liquidity, and what institutions should ask their counterparties in the meantime.
2. What the Binance breach revealed
The starting point is the scale and speed of the October 29 incident. In Jordan's telling, roughly $6.3 billion in digital assets moved out of Binance's hot wallets across 10 blockchains within 20 minutes. He describes it as the largest theft the industry has recorded.
He also acknowledges how the exchange responded. Binance paused withdrawals, drew on company funds to cover the losses its clients suffered, and brought its services back within days. Jordan does not dwell on that recovery, however. For him the episode matters less for how it was resolved than for what it exposed about where institutional money sits.
The issue he identifies is the omnibus hot wallet. When client assets are pooled in wallets that an exchange controls and keeps connected for fast withdrawals, a single successful breach can reach billions of dollars in minutes. A hot wallet is, by design, able to sign transactions quickly; pooling many clients' holdings behind that capability concentrates the damage any one failure can do.
Jordan frames this as a question of market structure in which the entire industry has a stake, rather than a lapse specific to one company. The point of the framing is that the exposure exists wherever client funds and trading infrastructure share the same custody arrangements, whether or not an attack has yet occurred.
3. Trading and safekeeping as separate disciplines
Exchanges, in Jordan's account, are essential infrastructure. They bring buyers and sellers together, supply liquidity and allow prices to form. Institutions need them, and nothing in his argument suggests otherwise.
Looking after large pools of client assets is, he contends, a different discipline with different demands. He lists four: segregation of one client's assets from another's, governance over how decisions about those assets are made, management of the cryptographic keys that control them, and legal protection for the client if a counterparty fails. Each of those is a specialism in its own right, and none of them is the core function of a business built to match orders at speed.
Traditional markets, he notes, settled this question decades ago. Someone who invests in equities places orders on an exchange, but the shares themselves are held by a custodian and transfers of ownership are settled through a central securities depository. Three separate institutions perform three separate jobs. Jordan argues that this division of labor is a major reason institutions are willing to trust those markets with large sums: a failure at the trading venue does not put the underlying assets within reach.
4. Why crypto developed a different model
Digital asset markets took another path for practical reasons, according to the column. In the early years, exchanges held customers' assets because there were few alternatives. The users of that period were mostly retail traders, and what they valued most was speed. Keeping funds on the platform meant being able to trade immediately.
The investor base has since changed. Jordan names pension funds, asset managers, banks and corporate treasuries among the organizations now allocating to digital assets. These institutions do not operate the way early retail traders did. They owe fiduciary duties to the people whose money they manage, they answer to internal risk committees, and they are supervised by regulators who want to know where assets are located and who has control over them.
An arrangement that was a reasonable convenience for an individual trader becomes a harder proposition for an organization that must justify its custody arrangements to a board or a supervisor. That gap between how the market was built and who is now using it is the tension running through Jordan's argument.
5. What is off exchange settlement?
Off exchange settlement is an arrangement in which an institution trades on an exchange while its assets stay with a separate, regulated custodian. Jordan presents it as the mechanism that makes the separation of trading and custody workable in practice.
Under the model he describes, the exchange can see the collateral the custodian is holding and can extend trading lines against it. The institution trades as it normally would. The assets themselves, however, move only when trades are settled. Between settlements they remain in custody, outside the exchange's wallets, so a compromise of the venue's own systems does not automatically expose them.
The practical effect is to change what an institution is exposed to when it deals with a trading platform. Where assets are deposited on the venue, the client depends on the exchange both to execute its orders and to keep everything it has handed over safe. Where assets stay with a custodian, the exchange's role narrows to execution, and the safekeeping risk sits with a firm whose business is organized around segregation, key management and governance, the requirements Jordan lists earlier in the piece.
Jordan says this has become the standard model for institutional access to centralized exchanges. In his view the question is no longer whether such a structure can function, but whether institutions that have not yet adopted it will review their arrangements in light of recent events.
6. The benefit he sees for exchanges
The column does not present custody separation as a loss for trading venues. Jordan argues that exchanges gain from it as well. Freed from the task of safeguarding large client balances, a venue can concentrate on what he regards as its real strengths: deeper liquidity, broader access and better execution.
There is also a commercial incentive. Some allocators cannot trade at all unless a regulated custodian is holding their assets. An exchange that supports off exchange settlement becomes available to those investors; one that does not is closed to them regardless of the quality of its markets.
Jordan notes that leading exchanges have already adopted the model for this reason. He adds that the venues which have done so are well placed to capture the next wave of institutional capital entering the sector.
7. Five questions for institutions
For organizations reassessing their setups after the Binance incident, Jordan offers a short checklist. The first question is where the assets are held and in whose name. Legal title matters most when something goes wrong, and an institution should know whether its holdings are recorded as its own or as part of a venue's balance.
The second is whether the assets are segregated or pooled with the holdings of other clients. The third asks what happens if the counterparty fails, which goes to the legal protection he lists among the requirements of proper custody.
Those first three questions map directly onto the weakness Jordan sees in omnibus hot wallets. Pooled holdings under a venue's control are the arrangement in which a single breach reaches the most money, and they are also the arrangement in which a client's claim is hardest to distinguish from everyone else's if the venue itself runs into difficulty.
The fourth concerns control: who is able to authorize withdrawals, and what checks apply before one is approved. The fifth is whether the institution can trade without first moving its assets onto the venue at all. Taken together, the questions amount to a test of whether an institution's exposure to a trading platform is limited to the trades it has open or extends to everything it owns.
8. Convergence with traditional finance
Jordan closes by placing the custody debate within a wider trend. Traditional finance and digital assets, he writes, are converging. Tokenized funds are launching on public blockchains, banks are building digital asset services into their core systems, and major jurisdictions are setting out clearer rules for the sector.
As that happens, he argues, the structure of the market has to look familiar to the institutions arriving in it. Firms accustomed to custodians and depositories standing apart from exchanges will expect to find an equivalent arrangement before they commit significant sums.
Separating trading from custody is, in his words, one of the clearest steps available. His conclusion is that doing so gives institutions the confidence to commit capital at scale, and that the October 29 breach has made the case more pressing than it was before. The column offers no timetable and announces no new product; its purpose is to argue that a model already in use should become the default.
