OnlyFans and the Channel Islands: An AML Case Study in Not Jumping to Conclusions

I heard about this the way most interesting things get discussed here: in a pub. Someone mentioned, half as a joke, that Jersey and Guernsey apparently punch above their weight on OnlyFans spending.

• It got a laugh.
• It also got me thinking about how you can't quite switch off after years working in financial crime risk: if that's true, what would it mean?
• Is it a curiosity, or is it a signal about something closer to home how money moves, how it's misused, and how the Islands are perceived by people who don't know them well?

That question turned out to be a better prompt than the original claim.

• What follows isn't really about OnlyFans.
• It's about what to do, analytically, the moment a headline number looks alarming.

INTRODUCTION

• A recent global ranking placed Guernsey 4th and Jersey 13th in the world for OnlyFans spending per 10,000 residents:
• Guernsey at roughly £62,300,
• Jersey at roughly £47,000.
• For two small international finance centres that live and die by their reputation for financial probity, that is exactly the kind of statistic that gets forwarded around a compliance team with a raised eyebrow.
• It's also, on its own, close to meaningless. Here's why, and here's the method for figuring that out, which is more useful than the answer.

THE REFLEX, AND THE DISCIPLINE

• The instinct, once you see a ranking like that, is to reach for the nearest evidence of harm:
• OnlyFans has been linked to money laundering elsewhere; therefore, a high per-capita ranking here means elevated risk here.
• That instinct isn't wrong to have. It's wrong to stop at.
• Good AML thinking:
• Isn't "does this look concerning";
• It's "what would I need to know before I could say whether it's concerning."
• Three questions do most of the work:
• Intensity, or prevalence? Is the number measuring concentration per capita, or how widespread the activity actually is?
• Transactional data, or modelled data? Was this measured directly, or inferred and estimated, and what can't the estimate tell you?
• Does evidence from elsewhere transfer? Documented misuse in one jurisdiction shows the harm is real somewhere. It doesn't, on its own, establish a base rate here.

3 QUESTIONS TO ASK

Question one: INTENSITY, OR PREVALENCE?

• Per-capita rankings measure intensity relative to population size, not the scale of activity.
• Guernsey has around 64,600 residents.
• Jersey around 104,100.
• Run the ranking's own figures through those populations, and the total annual spend implied is roughly £400,000–415,000 for Guernsey and £490,000–520,000 for Jersey combined, under £950,000 a year across two jurisdictions that each host tens of billions of pounds in banking and fund administration.
• This isn't a quirk of OnlyFans.
• Small, high-disposable-income populations mechanically produce extreme results on almost any per-capita consumer spending metric; the same effect shows up in league tables for
• Wine imports,
• Private jet charters, or
• Luxury car ownership.
• A "top five in the world" ranking can coexist with a total sum that rounds to noise against the sector it's supposedly worrying.
• Before treating any per-capita statistic as a risk signal, ask what the absolute number looks like once you multiply back out and what it looks like against the thing you're worried about protecting.

Question two: TRANSACTIONAL DATA, OR MODELLED DATA?

• The ranking in question isn't built from platform transaction records. It's modelled on Google Ads search-volume data, calibrated against OnlyFans' audited global revenue as a reasonable estimate, but one inferential step removed from the platform's own ledgers. It doesn't tell you:
• The number of users,
• The number of paying subscribers,
• The distribution of spend, or
• What share of the adult population is involved.
• A total of roughly £400,000 in Guernsey could be:
• A few hundred people spending heavily, or
• Several thousand people spending very little.
• Those are different stories with different implications, and the ranking can't distinguish between them.
• The general principle:
• Know whether the number in front of you was measured or inferred and
• Know specifically what it cannot tell you.
• A modelled estimate can still be useful, but it answers a narrower question than it appears to.

Question three: DOES EVIDENCE FROM ELSEWHERE TRANSFER?

• This is the question most likely to get skipped, because it's the one where a plausible-sounding link can feel like evidence when it's actually just a story, for example:
• Platform X has been used for financial crime; therefore, high spend on platform X here is worrying.
• Real documented misuse does exist, but it's worth being precise about what "documented" looks like, because it sets the bar the rest of the analysis has to clear.
• In February 2026, Turkish prosecutors detained at least sixteen people and seized assets worth roughly £5.1 million: companies, properties, vehicles, bitcoin, and gold in a probe that treated OnlyFans earnings (the platform is banned in Turkey) as proceeds of obscenity offences subsequently laundered through investment.
• In 2026, a St Louis man was indicted on eight counts of wire fraud after allegedly using power-of-attorney authority to take more than £545,000 from a retired teacher, spending part of it on OnlyFans purchases alongside property and personal expenses.
• In 2023, a whistleblower complaint filed with the US Treasury's financial crimes unit, made public by Reuters in January 2025, alleged that Mastercard and Visa had for years processed payments connected to child sexual abuse material and trafficking content on the platform.
• In 2023, in the US, a Florida personal-injury lawyer was charged with stealing more than £623,000 from sixteen clients, with bank records showing over £22,000 of it spent on OnlyFans.
• Notice what all four of these have in common:
• Named individuals or institutions,
• Quantified sums,
• Court filings or formal regulatory complaints, and
• A specific predicate offence.
• That's what real evidence of misuse looks like.
• Now compare that evidentiary bar against the claim implicit in the original pub conversation that high per-capita spend in Jersey and Guernsey specifically indicates a local financial-crime problem.
• There is no comparable public record for that claim.
• Not weak evidence, no evidence.
• The Turkey, Florida and St Louis cases are real, but they're evidence that misuse of this kind happens somewhere, not evidence about what's happening here.
• Citing one jurisdiction's documented cases as if they established a base rate for another jurisdiction is a category error, and
• It's the single most common way an alarming ranking turns into an unfounded conclusion.

ONLYFANS? APPLYING THE THREE QUESTIONS TO JERSEY AND GUERNSEY

• Earlier there was a suggestion that:
• OnlyFans has been linked to money laundering elsewhere; therefore, a high per-capita ranking here [Jersey/Guernsey] means elevated risk here.
• Run the three questions against the original claim and the picture is straightforward.
• The absolute sums are modest against either Bailiwick's financial sector.
• The underlying data is modelled, not transactional, and can't speak to user numbers or concentration.
• No public cases of OnlyFans-related embezzlement, misuse of family funds, or laundering of criminal proceeds have been reported in either jurisdiction.
• Jersey and Guernsey both operate mature AML/CFT regimes with generally positive MONEYVAL assessments, and existing customer due diligence and suspicious-activity-reporting processes are already built to catch unusual or unexplained outflows.
• An adult-content platform doesn't need bespoke treatment beyond that general risk-based discipline.
• That's not a claim that the theoretical risk is zero.
• Any platform enabling rapid, private, recurring discretionary spending creates the possibility [risk] that someone with a costly habit takes money from an employer or a joint account, or that someone with criminal income finds it a convenient outlet; the Florida and St Louis cases show that this possibility is real in principle elsewhere.
• It's a claim that the evidence to elevate that possibility to a local concern doesn't currently exist, and that a ranking built on search-volume modelling isn't the evidence to do so.

THE PLATFORM VERSUS THE PERSON

• Everything above deals with the platform, and the jurisdiction is:
• OnlyFans, in aggregate, a Jersey and Guernsey problem?
• It deliberately doesn't answer a different, narrower question:
• Could a person in either Island be misusing entrusted funds, and would a firm notice?
• That's the level the Florida and St Louis cases really operate at, and the ranking's answer and this question's answer don't depend on each other.
• The lawyer who stole from his clients and the man who took money from a retired teacher under power of attorney weren't caught because anyone was watching OnlyFans spending specifically.
• They were caught the way this kind of theft is usually caught: a client noticed money missing, a bill went unpaid, a pattern of outflows didn't match a stated income or an authorised purpose.
• OnlyFans was where some of the money went; it wasn't how the theft was detected, and it wouldn't need to be.
• That points to the actual, practical takeaway for firms here, and it has nothing to do with the ranking:
• Anyone holding client money, acting under power of attorney, or managing funds on behalf of a vulnerable person a trust and corporate service provider, a fiduciary, a professional adviser should already be alert to the same typology regardless of which discretionary platform the funds eventually land on.
• Red flags: outflows inconsistent with a client's known circumstances, spending that doesn't match a stated purpose, an authorised representative's personal expenditure pattern shifting after they gain access to someone else's funds.
• Existing CDD and transaction-monitoring standards are built to catch exactly that, and they don't need an OnlyFans-specific trigger to do it.
• The point isn't that firms should start watching for this platform by name; it's that the underlying vigilance the Florida and St Louis cases illustrate is already supposed to be running, independent of any ranking, and independent of which website the money happens to end up on.

A SECOND EXAMPLE: NATIONALITY IS NOT A MATCH

• The OnlyFans ranking was the first example:
• A curiosity, low stakes, easy to laugh off once you've run the three questions.
• The same category error shows up in a much more consequential setting, so it's worth running the same three questions against it.
• Take a customer born in Iran who has lived in London for twenty years, holds a UK passport, and
• Has no connection to a sanctioned entity, individual, or government body.
• Some firms, faced with that country of birth on a form,
• Quietly treat the file as elevated risk in a way that edges toward "sanctioned by association" extra friction,
• Delayed onboarding, or
• An unspoken assumption that greater scrutiny is inherently justified.
• The same pattern shows up with
• Customers connected to FATF grey-listed jurisdictions,
• Where residency or nationality alone gets read as a proxy for a propensity to launder money.

IN THE SECOND EXAMPLE: RUN THE 3-LEGGED FRAMEWORK

• Intensity, or prevalence?
• A sanctions designation
• Is a named match against a specific list entry: a person, an entity, a vessel.
• Country of birth is neither.
• Treating birthplace
• As if it carries the same evidential weight as a designation conflates a population-level characteristic with an individual one,
• is the same mistake as reading a per-capita spending ranking as if it described a specific person's behaviour.
• Transactional data, or modelled data?
• A sanctions match is about as transactional as data gets: a name, a list, a hit or no hit.
• FATF grey-listing, by contrast, is a country-level assessment of a jurisdiction's AML/CFT regime:
• Its laws, supervision, and enforcement record.
• It says nothing measured about any individual customer from that country, in the same way a search-volume model says nothing measured about any individual OnlyFans user.
• Grey-listing is a real, relevant data point for risk-based due diligence; it is not itself evidence against a person.
• Does evidence from elsewhere transfer?
• Documented laundering typologies connected to a jurisdiction: real cases, real patterns tell you something happens there.
• They don't tell you this customer, in front of you, with this transaction history, is doing it.
• That's precisely the leap the ranking encouraged with Jersey and Guernsey, applied here to a person instead of a place.

ENHANCED DUE DILIGENCE

• None of this argues against Customer Risk Assessments [CRA] and enhanced due diligence [EDD] where the facts warrant it, or against treating FATF status as one legitimate input into a risk-based approach:
• It's a well-established part of the toolkit for good reason.
• The argument is narrower and, I'd suggest, more important:
• Nationality or birthplace, on its own, is not a match, not a red flag, and not a substitute for looking at the actual customer and the actual transaction.
• Firms that skip that step
• Aren't being cautious.
• They're doing the version of this article's opening mistake that has real consequences for real people, rather than for a headline.

THE TRANSFERABLE POINT

• The next headline number won't be about OnlyFans, and it won't always be as low stakes as a spending ranking. It might be a crypto-exposure statistic, a grey-list update, or a sanctions story that tempts one to take a shortcut.
• The three questions don't change:
• What's actually being measured,
• Is the underlying data real or modelled, and
• Does the supporting evidence actually transfer to the claim being made.
• A pub conversation is a fine place for a headline number to start a thought. It's not a good place for it to end, and neither is a league table, a list, or a passport on its own.

End

DISCLAIMER

The views in this article are my own and do not represent those of any employer, past or present. I'm not a lawyer, and nothing here is legal or regulatory advice.

I aim to start a conversation about how we think about and manage risk. I'd welcome pushback.

Mathew Beale - mathew@comsuregroup.com

SOURCES

*PLEASE NOTE:

• I haven't independently verified the OnlyGuider ranking or World Population Review page beyond taking them at face value as the origin of the headline stat; that's the one piece of source material in here I couldn't cross-check against a second independent outlet, which is worth noting since it's the number the whole piece hangs off.
• The Turkey case is a live investigation, not concluded; therefore, the story may have shifted by the time the article is read for example, whether charges have been formally filed or dropped, since “detained” and “seized assets” can change.

The ranking:

• OnlyGuider, “OnlyFans Wrapped 2025”
• https://onlyguider.com/blog/onlyfans-wrapped-2025/
• World Population Review, “OnlyFans Spending by Country”
• https://worldpopulationreview.com/country-rankings/onlyfans-spending-by-country

Turkey investigation (February 2026):

• Reuters, via Froggy 99.9 KVOX, “Turkey detains 16 in OnlyFans probe, seizes $6.9 million in assets”
• https://froggyweb.com/2026/02/13/turkey-detains-16-in-onlyfans-probe-seizes-6-9-million-in-assets/
• Turkish Minute, “Turkey targets 25 suspects for selling explicit content on OnlyFans, seizes $6.9 mln in assets”
• https://www.turkishminute.com/2026/02/13/turkey-targets-25-suspects-for-selling-explicit-content-on-onlyfans-seize-6-9-mln-in-assets/
• Hürriyet Daily News, “16 detained over money laundering claims tied to OnlyFans”
• https://www.hurriyetdailynews.com/16-detained-over-money-laundering-claims-tied-to-online-adult-content-218915

Florida lawyer case (2023):

• WFLA, “Pinellas County lawyer accused of stealing $840K from clients”
• https://www.wfla.com/news/pinellas-county/pinellas-sheriff-discusses-arrest-of-lawyer-who-allegedly-stole-thousands-from-clients/
• Fox News, “Florida attorney blew $840k in stolen client money on OnlyFans porn site, drugs: sheriff”
• https://foxnews.com/us/florida-attorney-blew-stolen-client-money-onlyfans-porn-site-drugs-sheriff.amp

St Louis power-of-attorney case (2026):

• First Alert 4, “Man charged with stealing $735,000 from retired St. Louis teacher”
• https://www.firstalert4.com/2026/05/14/man-charged-with-stealing-735000-retired-st-louis-teacher/
• Fox19/KMOV, “Man accused of stealing $735,000 from retired teacher, using money on OnlyFans”
• https://www.fox19.com/2026/05/15/man-accused-stealing-735000-retired-teacher-using-money-onlyfans/

Whistleblower complaint (filed 2023, reported 2025):

• Reuters, via US News, “Exclusive-U.S. Whistleblower Says Mastercard, Visa Failed to Stop Payments for Child Sex Abuse Material on OnlyFans”
• https://www.usnews.com/news/world/articles/2025-01-24/exclusive-u-s-whistleblower-says-mastercard-visa-failed-to-stop-payments-for-child-sex-abuse-material-on-onlyfans
• CBS News, “Mastercard and Visa accused of enabling payments for child sexual abuse content, report claims”
• https://www.cbsnews.com/news/mastercard-visa-onlyfans-child-abuse-fincen-whistleblower-reuters-allegation/

Currency conversions from USD to GBP use an approximate rate of £1 = $1.35 (as of July 2026) and are indicative only.

All figures should be independently re-verified before publication, particularly given how quickly rolling investigations (e.g. Turkey) can develop.