Invest1 distinct publisher2 min readUpdated
A recruiter's Hong Kong prison term stands, and the proceeds it fed ran on stablecoins. The gap between traced fraud flows and estimated global losses shows where screening is not looking.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
Set the file's two largest figures beside each other. The traced on-chain flow into fraud-linked wallets for 2025 comes to roughly 3.2 percent of the $442 billion INTERPOL attributes to global fraud losses that year [1][2].
That gap admits two readings, and Chainalysis picks one: it expects its own 2025 total to climb past $17 billion as further scam wallets are identified [4], a 21 percent upward revision to a number already in print [2]. Labelling trails the money. A risk model calibrated to the published aggregate is calibrated to a figure its author has said is too low.
Seizure numbers are lumpier than flow numbers. The DOJ indictment of Prince Group's Chen Zhi came with a Bitcoin seizure of around $15 billion [14], which exceeds the entire traced year of fraud wallet inflows by about 7 percent [4]. Value at that scale surfaces through case files, not wallet screening.
The account of stablecoin preference is mechanical: the tokens hold value in transit and convert into local currency through laundering networks operating in China [5]. The control point is the conversion, not the transfer, and conversion is where exchanges sit. INTERPOL's illustration of what arrives there is a 20-year-old suspect located in Thailand with more than $122.5 million in romance scam transactions over 10 months [9], about $12.25 million a month [5], routed through cross-chain swaps meant to obscure origin. Single-chain screening sees a fragment of that.
UNODC's 2026 report describes Southeast Asian syndicates as separate businesses sharing infrastructure, with most criminal profit laundered on blockchain networks [10]. Delphine Schantz, the agency's regional representative for Southeast Asia and the Pacific, calls the model "corporate franchising: imagine specialised departments for laundering money, trafficking people, smuggling migrants, and harvesting data" [11]. An onboarding desk meeting one of those laundering departments sees a client whose entire business is conversion volume, and nothing about the fraud that produced it.
The Hong Kong arithmetic is worth reading closely for what it prices. The court set a seven-year base and removed a third for the guilty plea [15]. Ma Che-hou, 32, admitted conspiracy to defraud and money laundering across 2021 and 2022 after persuading five men aged 20 to 32 with offers of well-paid work, business opportunities or online romance [6]; some ended up held at KK Park in Myanmar, where they were tortured with electric shocks [7].
The ledger also does not close. Moving 2025 from $14 billion toward $17 billion is last year being re-scored with this year's labels, and the same re-scoring will reach the transactions exchanges cleared while those labels did not exist.
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
Hong Kong's Court of Appeal confirmed a 56-month prison sentence for syndicate recruiter Ma Che-hou over fraud and money laundering linked to human trafficking.
INTERPOL estimates the global fraud economy caused losses of $442 billion in 2025.
Chainalysis tracked more than $14 billion sent to fraud-related crypto wallets in 2025.
Chainalysis expects its 2025 fraud wallet total to amount to over $17 billion as more scam wallets are uncovered.
Stablecoins such as USDT are a key laundering rail, offering fast cross-border transfers, preserving value, and converting easily into local currencies through money laundering networks operating in China.
Ma Che-hou, aged 32, confessed to conspiracy to defraud and money laundering in 2021 and 2022; prosecutors said he persuaded five men aged 20 to 32 with offers of well-paid jobs, business chances or online romance, and the men ended up in Southeast Asia.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Named third-party findings, single-outlet relay
The factual spine is specific and attributable: a citable judgment reference, sentencing arithmetic, a named UNODC official quoted directly, and numbered figures credited to Chainalysis, INTERPOL, DOJ and OFAC. But every item reaches the reader through one crypto trade outlet with no primary document, no second publisher and no subject response, and the prose is loose in places (for example 'cross-chain swipes'). The causal claim implied by the framing - that this prosecution's proceeds ran on stablecoins - is never evidenced for the case itself.
Quantified illicit flows met by billion-scale enforcement
Both sides of the story show real-world scale rather than pilots: traced fraud-wallet inflows above $14 billion for 2025 with an expected revision past $17 billion, an 85 percent rise in flows to trafficking services, and enforcement that has moved to production scale with a $701.9 million strike force seizure, 503 site takedowns, 29 OFAC designations and a roughly $15 billion Bitcoin seizure. What tempers the score is that adoption of the countermeasure side - actual screening changes at exchanges or stablecoin issuers - is recommended in the article but nowhere demonstrated.
Mildly overstated causal link
The numbers themselves are attributed and mostly conservative, and the article volunteers the deflating fact that public chains help investigators. The overstatement is structural: a Hong Kong recruiter case in which no crypto transaction is described is used as the entry point for a global stablecoin off-ramp thesis, and estimate-based loss totals ($442 billion, near $500 billion) sit beside traced on-chain totals without methodological reconciliation. Vendor-sourced flow figures and agency seizure headlines are relayed without challenge.
Vendor and agency framing relayed by trade press
The load-bearing statistics come from a commercial blockchain analytics vendor whose product addresses exactly the screening gap the article urges exchanges to close, and the enforcement figures come from agencies publicising their own seizures and designations, all carried by a crypto trade publication whose audience is the compliance-buying industry. None of these interests is disclosed in the piece. This is observable from the attribution pattern in the source and does not require imputing motives beyond it.
Moderate-low: specifics are checkable, corroboration is absent
Confidence is limited chiefly by cluster structure: one publisher, one article, no primary documents and no opposing voice, so single-source error would propagate undetected. It is lifted by the presence of falsifiable specifics - a case citation, a named quoted official, exact dollar and count figures, and arithmetic that reconciles internally - which makes the core court and enforcement facts likely to survive verification even if the stablecoin causal framing does not.
invest
DOJ Flew a $165M Ponzi Case Home From Fiji, Which Resets the Affiliate Math1 distinct publisher
security
China got the handcuffs on Chen Zhi. Expect the fraud to change address, not stop1 distinct publisher
invest
The card networks just picked the referee for agent checkout, and it looks like EMVCo2 distinct publishers
invest
Two years of stablecoin plumbing, and 0.2% of euro area online sellers take crypto2 distinct publishers
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 21, 2026