The $46 Million Crypto Theft Case Against John Daghita Exposes a Government Custody Failure

The Department of Justice’s March 2026 indictment of John Dean Daghita, filed in the Eastern District of Virginia, reads like a case about theft. It is a case about custody.

Federal prosecutors allege that Daghita, a contractor hired as a “Cryptocurrency Subject Matter Expert” to support the U.S. Marshals Service, misappropriated approximately $46 million in government-controlled digital assets and routed them into wallets he controlled. The indictment charges wire fraud, theft of government property, money laundering, and unlawful monetary transactions, and seeks forfeiture of both the crypto and the assets allegedly purchased with it.

Let’s start with this:

Custody is the entire game in digital assets.

The indictment alleges that a contractor entrusted with handling U.S. Marshals Service cryptocurrency moved approximately $46 million in government-owned assets into wallets he controlled.

That allegation lands with force because it reflects a failure at the only point that matters. Control of digital assets does not sit in policy. It does not sit in contracts. It sits in keys, and the systems that govern those keys.

Everything else is commentary.

Authority Was Concentrated Where It Should Have Been Broken Apart

The system placed influence, execution, and practical control in the same hands.

The indictment alleges that the defendant was responsible for executing transfers, swaps, and storage of government crypto assets, and then transferred those assets into external wallets he accessed outside the approved environment.

That structure collapses the moment you look at it with any seriousness. Financial systems survive by separation. Initiation sits in one place. Approval sits in another. Custody sits behind both.

When those lines blur, the system stops being a system. It becomes a person.

I have said this before in the context of digital assets and regulatory design. The law does not struggle with innovation. It struggles with concentration. Crypto simply exposes that truth faster and more brutally than anything that came before it.

Competence Was Performed. Control Was Never Proven

The justification allegedly offered for moving assets tracked industry best practices: segregation, multi-signature structures, wallet hygiene.

That is exactly why it worked.

The indictment alleges that the defendant told the U.S. Marshals Service he needed to move funds to safer wallet structures rather than commingle them.

That language carries authority. It signals expertise. It sounds like the right answer in the room.

But digital assets do not reward people who sound right. They reward systems that are right.

There is a difference between describing a multi-signature environment and operating one that no single actor can override. One is language. The other is architecture. Institutions that fail to distinguish between the two eventually learn the distinction the hard way.

The Timeline Shows a System Without Teeth

The system allowed escalation.

The indictment alleges an initial unauthorized transfer of roughly $5 million in December 2025, followed by approximately $41 million in additional transfers in January 2026.

That sequence should not exist in a functioning control environment.

A single anomalous transfer should trigger containment. Access should tighten. Permissions should contract. Independent verification should take over.

Here, the opposite allegedly occurred. The system remained permissive while the scale of movement increased.

That is not a failure of detection. That is a failure of posture.

Speed Eliminated the Possibility of Correction

Digital assets move with finality.

The indictment alleges that when the U.S. Marshals Service requested the return of approximately $2 million in cryptocurrency, the funds were redirected within roughly twenty minutes.

That is how this market works. Transactions settle when they are broadcast and confirmed. There is no institutional pause button.

I have written before that digital assets compress time in a way regulators and institutions still underestimate. Decisions that would unfold over days in traditional finance resolve in minutes here. That compression rewards preparation and punishes hesitation.

Controls that operate after execution do not function as controls. They function as documentation.

Once Assets Leave Controlled Custody, the Case Changes Entirely

The indictment alleges subsequent use of Tornado Cash to obscure the origin and ownership of funds.

At that point, the story shifts.

The question stops being “How do we prevent this?” and becomes “How do we trace this?”

Blockchain analytics can reconstruct flows. Law enforcement can seize devices, trace wallets, and build forfeiture cases. Companies like Zealous Markets or ZeroShadow often help.

The indictment itself seeks over $46 million and identifies specific crypto holdings, hardware wallets, and luxury assets tied to the alleged proceeds.

That is recovery work. It is necessary. It is often effective.

It is also downstream.

Custody exists to keep institutions from living downstream.

The Spending Is Loud. The Failure Is Quiet

The indictment alleges purchases that read like excess: a Lamborghini, luxury travel, private villas, a bodyguard, a private chef, and high-value digital assets including Telegram usernames purchased for hundreds of thousands and, in one instance, over $1 million.

That is the part people will remember.

It is also the least interesting part of the case.

Fraud looks obvious after the fact because spending creates a visual. The real story sits earlier, in the structure that allowed funds to move without friction, without interruption, and without independent control.

Every major financial failure eventually reduces to that moment.

The Government Is Operating at Scale Without Institutional-Grade Crypto Custody

The U.S. Marshals Service sits at the center of federal crypto asset forfeiture.

The indictment itself describes the Marshals Service as responsible for managing and disposing of seized cryptocurrency through its Asset Forfeiture Division and Complex Assets Unit.

That role carries enormous weight. Federal seizures have produced billions in digital assets over the past decade.

Scale demands rigor.

When an institution operates at that level, custody cannot be treated as a vendor function or a technical afterthought. It is the function. It is the balance sheet. It is the exposure.

And when that function is weak, the consequences scale with it.

This Case Lands on a Point the Industry Has Already Settled

Digital assets do not tolerate ambiguity around control.

Keys determine authority. Systems determine whether that authority is shared, constrained, or absolute.

The indictment alleges a structure where one individual could influence movement, execute transfers, and control destination wallets.

That structure produces a predictable outcome.

I have spent a significant amount of time thinking about how law and innovation intersect in this space. The conclusion has remained consistent. The law does not need to be reinvented to handle digital assets. It needs to be applied with discipline.

Segregation of duties. Independent authorization. Verifiable custody. Continuous reconciliation.

Those principles are not new. They are simply non-negotiable here.

Final thoughts

The market already answered the custody question.

Institutions either build systems that distribute control or they accept that control will concentrate somewhere they cannot see.

The indictment alleges that concentration existed here, and that it carried $46 million with it.

That is not a crypto problem.

That is a governance decision with consequences that settled exactly as designed.

That’s all for now,

Braeden

————————————————————

About the author

K. Braeden Anderson is a Partner at Gesmer Updegrove where he leads the firm’s Securities Enforcement & Investigations practice, and chairs Mackrell International’s Blockchain & Digital Assets Group and Securities Enforcement & Investigations Group. He is a nationally recognized securities regulatory and enforcement attorney whose practice sits at the intersection of traditional financial regulation and emerging technology. He has been recognized in Best Lawyers: Ones to Watch® in America (2025) for Financial Services Regulation Law and Securities Regulation, and was named the #1 most-read fintech thought leader in the United States in Mondaq’s Spring 2025 Thought Leadership Awards.

Before joining Gesmer Updegrove, he served as Assistant General Counsel at Robinhood Markets, Inc. (NASDAQ: HOOD), advising on high-stakes regulatory and enforcement matters, and earlier practiced at Kirkland & Ellis LLP and Sidley Austin LLP in New York and Washington, D.C.