A money mule is a person whose bank account is used to receive and move criminal proceeds, whether or not they understand what the money is. A mule account is that account. Both terms describe a role in a laundering chain.
A student gets a message in a WhatsApp group: a payment-processing job, a few thousand rupees a week, just receive transfers and pass them on. Nothing in it mentions crime. Six weeks later the account’s frozen, there’s a police complaint, and the money has gone through four more accounts in a chain nobody can unwind.
Let’s cover how banks can catch mule accounts without freezing the genuine customers who look like them on paper.
What a money mule is
A money mule moves money that belongs to a crime. It makes no difference whether the account holder was talked into it by a recruiter, pressured by someone they know, or paid knowing exactly what the account was for.
Almost everything else about mule activity follows from the fact that criminals need bank accounts in other people’s names.
The definition, and who it covers
Mules fall into two groups, and the two need different responses from a bank.
Unwitting mules believe they are doing something legitimate. The recruitment arrives as a part-time job, a payment-processing role, or a favour for a partner met online. Complicit mules know the money is dirty and rent out the account anyway, which is now a market with its own pricing.
The Reserve Bank of India defines a mule account as a bank account used by criminals to launder illicit funds, often set up by unsuspecting individuals lured by promises of easy money or coerced into participation.
The Financial Action Task Force, the intergovernmental body that sets global anti-money-laundering standards, describes money mules more broadly as people used to transfer value, either by laundering stolen money or by physically transporting goods. Both definitions are collected in PwC India’s 2025 analysis of mule fraud.
Both definitions cover the willing participant and the deceived one. A bank cannot tell them apart at the point of detection anyway.
What a mule account is not
Three other fraud types get filed under the same heading, and each needs a different control, so the distinction is operational rather than academic.
| Term | What it is | Whose money |
|---|---|---|
| Mule account | An account used to receive and forward criminal proceeds | Someone else’s, stolen from a victim |
| Smurfing | Breaking a large sum into deposits small enough to stay under a reporting threshold | The criminal’s own, being placed |
| First party fraud | A customer using their own real identity to defraud the institution | The institution’s |
| Bust out fraud | Building good credit on an account, drawing it down fully, then disappearing | The lender’s |
First party fraud and bust out fraud involve a real customer acting against the bank with their own name attached. A mule account involves a real or fabricated customer acting as a conduit for someone else’s victim. Synthetic identity fraud cuts across all four, because a fabricated identity can be put to any of them.
Smurfing is the one that gets mixed up with mule activity most often. Structuring deposits keeps a sum under a reporting threshold. Muling puts a stranger’s name between the criminal and the money.
How does a mule account get opened?
Every mule account begins with an account opening, and that is a moment a bank controls completely. Monitoring starts later and catches less.
Three routes account for most mule supply, and they are not equally hard to stop.
Three ways in
- A real person is recruited and verified as themselves. The documents are genuine, the face matches, and every check passes, because nothing about the application is fake. What is fake is the purpose. This is the only route that leaves a prosecutable person at the end of the chain.
- A stolen identity is used to open an account the victim never sees. Real documents belonging to someone else, often harvested in bulk. The victim finds out when a police notice arrives.
- A synthetic identity is assembled to pass verification. A fabricated person built from a mixture of real and invented attributes, with a face to match. Nobody is defrauded at the opening, so nobody reports it.
Route one is a human problem and needs awareness campaigns and behavioural monitoring. Routes two and three are identity verification problems, and they are the ones that scale, because a recruiter can sign up a few students a week while a script can attempt hundreds of applications a night.
Mule accounts opened with deepfake KYC
Remote onboarding removed the branch visit, so the face in the video call or the selfie capture is now doing the work a bank officer used to do.
HyperVerge builds AI-powered identity verification for digital onboarding, with fraud prevention and deepfake detection in the same flow. That check at sign-up is where a mule account either gets opened or gets stopped.
Two attack shapes matter here. In a presentation attack, something is held up to the camera: a printed photo, a screen, a mask. In an injection attack, synthetic video is fed into the application past the camera entirely, through a virtual camera or a capture device.
The second is what turns mule supply into a volume business. A recruited student is one account. A generated face driven through an injected feed is as many accounts as the attacker has documents to pair it with, and it never has to be paid, never changes its mind, and never talks to the police.
Video KYC flows in India are where this collides with regulation most directly, because the RBI treats the capture session itself as part of the control environment rather than as a formality.
“Mule accounts have moved to the top of the regulator’s agenda, and that has changed what banks ask us for. The question used to be whether an account was opened by a real person. Now it is what happens in the days after one gets flagged, when a bank has to decide whether it is looking at a launderer or at a customer who will be locked out of their own salary for a week. That investigation is what we are building EDD AI for.”
Kedar Kulkarni, CEO, HyperVerge
The red flags, at two different moments
Mule detection has two windows, and the industry publishes advice about only the later one. An account opening generates evidence, and a first suspicious transfer generates evidence, and they are different evidence.
Reading both is the difference between finding a mule network and finding one account from it.
Red flags at account opening
None of these requires a transaction to have happened yet. They are visible in the application itself, and most of them only become visible when applications are compared against each other rather than assessed one at a time.
- The same device fingerprint appearing across applications in different names.
- A selfie that matches a face already enrolled under a different identity.
- Documents reused with small edits: same template, same scan artifacts, different numbers. Document forgery signals and fake ID detection both land here.
- Address, phone or email clustering, where dozens of unrelated applicants share a contact point or a building.
- A sudden burst of applications from one geography, one device model, or one time window.
- An applicant profile that does not fit the product: a first-time account holder opening a high-limit current account with no business history.
From inside a live onboarding queue these cluster in a way that’s obvious in aggregate and invisible one application at a time. That is the practical argument for comparing applicants against a history rather than scoring each one alone, and it is why enhanced due diligence triggered on a single application rarely catches a network.
Red flags once the account is live
The Indian Banks’ Association sets out the features that characterise a mule account in operation: frequent international transfers, an unusually high number of counterparties, rapid transaction velocity, and sudden surges in account activity.
Each one describes the same underlying behaviour: money arriving and leaving fast, to many places, in a pattern that has nothing to do with how the account holder lives. A salary account that takes one credit a month and then takes forty in a week isn’t a salary account any more.
These are the signals that trigger a suspicious transaction report, and they do work. They just work after the money has moved.
Warning signs for a person being recruited
If you are being offered money to receive payments and pass them on, the offer is the warning sign. A few specifics worth knowing:
- A job that pays well for almost no work, especially one advertised through social media or a messaging app.
- Any request to use your personal account for someone else’s business. Legitimate employers do not route company money through a new employee’s savings account.
- Instructions to move money onward to accounts you were not told about in advance.
- Pressure to keep the arrangement quiet, or to act quickly.
Not knowing the money was stolen isn’t automatically a defence, and the account gets frozen either way.
Where the law actually stands
India has no offence called “being a money mule”. Mule activity is prosecuted through general money laundering law, which shapes how banks and police can respond and explains a good deal about why response is slow.
That gap is not unique to India, and the comparison is instructive.
India
Money mule activity is addressed through the Prevention of Money Laundering Act 2002 and related laws rather than through a dedicated offence. The RBI’s KYC Master Direction requires banks to monitor and identify suspicious accounts, including those operated by money mules, and to report them to FIU-IND, India’s financial intelligence unit, through suspicious transaction reports.
PwC’s 2025 analysis records that these frameworks support detection and reporting but do not explicitly criminalise money mule operations, with penalties under PMLA running up to seven years.
On the detection side, the position has moved quickly. MuleHunter.AI, built in-house by the Reserve Bank Innovation Hub, uses machine learning to identify suspected mule accounts, and the Government of India’s Press Information Bureau records it as live in 26 banks and being scaled up further.
Banks have separately been advised to deploy real-time transaction monitoring and network analytics, so the target is a mule network rather than one account at a time.
That bank count is a point-in-time figure from a government statement and it has been rising, so check the current number before quoting it anywhere that matters. HyperVerge’s breakdown of RBI video KYC and deepfake guidelines covers the onboarding side of the same regulatory picture.
How five jurisdictions compare
PwC’s 2025 report sets out how five jurisdictions treat mule activity, and the split between those with a dedicated offence and those without is sharper than the penalty ranges suggest.
| Jurisdiction | Explicit mule offence | Framework | Penalties |
|---|---|---|---|
| UK | Yes | Proceeds of Crime Act, Theft Act, Serious Crime Act | Up to 14 years |
| Australia | Yes | AML/CFT Act 2006, AUSTRAC, Criminal Code | Up to 25 years |
| USA | No | Fraud statutes and money laundering laws | Up to 30 years, fines above $1 million |
| EU | No | EU AML Directives and national law | Institutional fines of €5 million or 10% of turnover |
| India | No, via PMLA | PMLA 2002, RBI KYC and AML rules | Up to seven years, fines and confiscation |
A jurisdiction without an explicit offence isn’t a soft one. The US carries the longest sentences here and has no dedicated mule offence at all. What the explicit offence changes is how early an investigator can act, because charging someone with operating a mule account is simpler than proving their part in a laundering scheme.
What can banks actually do about mule accounts?
Mule controls work in sequence, and the order matters more than the individual tools. Catching a mule account at the first suspicious transfer is a worse outcome than not opening it.
Five steps, in the order they bite.
Close the opening, then watch the account
Mule controls run in sequence, and the sequence is what most programmes get backwards. Every step below is cheaper than the one after it, and the first two are the only ones that run before an account exists.
- Verify identity properly at opening, including liveness and document authenticity rather than a document photo alone. This is where routes two and three from the opening section get stopped.
- Compare the application against history as well as against the applicant’s own documents. Device, face and contact-point reuse across applications is the signal that finds networks.
- Risk-score the account at opening and carry that score forward, so an account that looked unusual on day one is monitored differently on day ninety.
- Monitor for the operational red flags: velocity, counterparty count, international transfers, activity surges.
- File the report and act on the outcome. An STR that goes out without an internal action attached catches nothing.
Where identity verification earns its place here is steps one and two. Passive liveness and face spoofing checks address what arrives at the camera; a liveness check at capture is the usual anchor. Neither replaces monitoring, and anyone selling it that way is overselling it.
Money mule fraud is a supply problem and a movement problem at the same time, and AML compliance programmes that only address the second half will keep finding accounts after the money is gone.
What to do when an account is already flagged
If your account has been frozen in connection with suspected mule activity, contact the bank branch directly and ask what is required, then cooperate with the investigating authority rather than waiting it out.
Banks cannot usually unfreeze an account flagged through a law enforcement process on their own authority. Provide the source of the credits you received, keep records of the communications that led to the arrangement, and get legal advice if there is a police complaint. Opening another account elsewhere while the first is under investigation makes things worse, not better.
Shortening the investigation that follows a freeze
A freeze is not a verdict. Once an account is flagged, someone at the bank has to work out whether this is a launderer or a customer caught by a pattern, and that decision is enhanced due diligence.
Done the traditional way it takes seven to ten days. A background check is ordered from one provider, a field agent goes out to confirm the address and the person, a letter goes to the customer, and somebody reconciles the pieces by hand.
That week costs both sides. For a launderer it is seven days of money sitting safely out of reach. For an honest customer it is seven days with no salary, no UPI and no way to pay the rent.
HyperVerge’s EDD AI engine runs the same investigation as a set of coordinated AI agents, with a human still signing off. Background due diligence covers criminal records, adverse media, sanctions and business filings. A video field investigation replaces the physical visit, confirming location, liveness and face match and walking the premises on camera.
Location intelligence reads the address in context: what sits within a kilometre of it, and whether that fits the account being opened. A decision lands in minutes instead of days, which protects the honest customer and the bank at the same time.
Mule supply is a front-door problem that most programmes treat as a monitoring problem. Talk to our team about what identity verification surfaces at account opening, and where EDD AI fits once an account is already frozen. Or book a demo and put your own onboarding queue through it.
FAQs
What is a money mule?
A money mule is a person whose bank account is used to receive and forward money that came from a crime. The Reserve Bank of India defines a mule account as one used by criminals to launder illicit funds, often opened by people lured with promises of easy money or coerced into participating.
How many mule accounts are there in India?
There is no single published total. The figures available come from different counting bases: accounts flagged to banks, accounts frozen, and accounts detected by individual institutions. Quote the specific measure and its date rather than a headline number, because the three are routinely mixed up.
What is a key red flag for identifying a money mule account?
The Indian Banks’ Association names four: frequent international transfers, an unusually high number of counterparties, rapid transaction velocity, and sudden surges in activity. The strongest single signal is money arriving and leaving quickly, to many destinations, in a pattern unrelated to how the holder normally banks.
What are the punishments for money mule in India?
India has no dedicated money mule offence. Activity is prosecuted under the Prevention of Money Laundering Act 2002 and related laws, with penalties running up to seven years along with fines and confiscation, according to PwC’s 2025 analysis. The account is typically frozen well before any charge.
How do I remove a freeze from my bank account?
Contact the branch and ask what the freeze relates to, then cooperate with the investigating authority. A bank usually cannot lift a freeze placed through a law enforcement process on its own authority. Keep records of the credits received and the communications that led to the arrangement.
Can a deepfake be used to open a bank account?
Yes, and remote onboarding now tests for it. A generated face can be presented to a camera or injected into the application past the camera. Liveness detection and injection attack detection are the controls aimed at each, and they are different checks.



