impersonation fraud

basicwidespread

A type of fraud where the perpetrator pretends to be a trusted person or institution – nowadays often using AI-forged voice or face – to extract money or data.

In plain terms

The essence of impersonation fraud is that someone pretends to be another person: your bank's customer service representative, your boss, your child, a courier company, or a public authority. They do not hack your computer but exploit your trust: they want you to transfer the money or hand over your password yourself.

It is like someone ringing the doorbell in a postman's uniform: because of the uniform, it does not occur to you to ask if they are really a postman. The scammer's uniform used to be an official-looking email or a confident phone voice; today, it is increasingly prepared with the help of artificial intelligence.

Due to AI, this type of fraud has become cheaper and more convincing. A person's voice can be cloned from a few seconds of audio recording, a moving, talking video can be created from one or two photos, and language models write messages in flawless Hungarian, personalized. The old warning signs, poor spelling and strange phrasing, no longer protect you.

How it works

Data collection: the scammer gathers information from social media, corporate websites, or leaked databases about who you are, who you are connected to, and who is worth impersonating in front of you.
Creating a disguise: fake caller ID, deceptive email address, cloned website, or AI-cloned voice or deepfake video of the person being impersonated.
Contact with urgency: they call, message or start a video call and create an emergency ("there is a suspicious transfer on your account", "the bill must be paid immediately", "mom, I'm in trouble") so that you have no time to think.
The requested step: persuades the victim to transfer money to a "security account," dictates the confirmation code, installs a program granting remote access, or hands over login credentials.
Making the money disappear: it is routed through several accounts or converted to cryptocurrency within minutes, so it is very hard to recover afterwards. Part of this process can now also be carried out by AI agents, which write to and phone people at scale, yet in a personalized way.

An everyday example

Your phone rings, the display shows your bank's name, and a polite 'security officer' on the line claims someone is trying to withdraw money from your account, so you must urgently transfer your savings to a protected account. In such cases, hang up and call the bank yourself using the number on the back of your card: a real bank never asks for this.

Who is affected

everyoneAnyone can be called in the name of their bank, an authority, or a family member, and the forged voice or phone number can make the call appear completely legitimate at first glance.

bank customer, investorThe most common scenario is a fake bank call and a fake investment promoted with deepfake videos of celebrities, where the victim themselves approves the transfer. → Money

entrepreneur, executiveAn 'urgent transfer request' arriving in an executive's name, even in their cloned voice, can cause serious damage to the company with a single call.

employeeWorkers in finance, HR or customer service are targeted most often, with fake-boss, fake-supplier or fake-IT approaches. → Work & HR

A common misconception

MythOnly elderly or non-tech-savvy people fall for such scams.

In realityA well-structured, AI-forged request has already deceived trained financial experts and IT specialists. It does not rely on ignorance, but on urgency, deference to authority, and helpfulness.

Limits and risks

Voice, face, displayed phone number, and email address no longer prove on their own that you are speaking with the person they claim to be.
Deepfake detection programs are not fully reliable, so the best defense is a callback on a number you know independently, and within the family, a pre-agreed secret code word.
If you personally approved the transfer, recovering the money is difficult and uncertain, making prevention far more valuable than post-event complaints.
Urgency is almost always a telltale sign: if someone demands an immediate transfer, code, or confidentiality, stop and verify via another channel.

Similar, but not the same

phishing – Phishing extracts data through fake messages and websites, mostly in bulk, sometimes targeted, and often impersonating a well-known company. Impersonation fraud is a broader term: it covers any fraud based on pretending to be a trusted person or institution, via any channel, including phone or video calls.
deepfake – Deepfake is the technology used to create fake audio or video; impersonation fraud is the crime for which it can be used as a tool, but it can also be committed without deepfakes.
Identity theft – In identity theft, others act in your name using your data (for example, taking out loans); in impersonation fraud, they pretend to be someone else in front of you.

Related concepts

Broader

psychological manipulation, online fraud, cybercrime

Narrower

voice cloning fraud, CEO fraud, grandparent scam, fake bank call

On the edge

deepfake, phishing, voice cloning, Identity theft, Phone phishing

A brief history

Extracting money by impersonating others is an ancient method; for example, the "grandparent" phone scam has been known for decades. The first widely known case in the AI era occurred in 2019: reports state that a British energy company executive was induced to transfer approximately 220,000 euros by a voice synthesized to sound like the CEO of the German parent company. In 2024, an employee at the Hong Kong office of the British engineering firm Arup transferred an amount equivalent to approximately 25 million US dollars to scammers in several installments following a deepfake video conference.

Where things stand

As of late 2026, this is one of the most common types of fraud: voice cloning and live deepfake video have become cheap tools available to anyone. Banks, telecommunications companies and authorities defend themselves with screening, verification steps and awareness campaigns, but scammers' methods change rapidly.

Status: 2026 November

In the news

Appears in 9 Agent Age stories; the curve shows weekly occurrence in the news magazine.

88-year-old loses $450,000 in deepfake scam; bank will not reimburse under Canadian rulesNov 8, 2026 · Money
Only one in five people in Hong Kong trust banks to stop deepfake scamsNov 1, 2026 · Money
Deepfake scammers use faces and voices of central banks and officials in Romania, Malaysia and IndiaOct 19, 2026 · Money
FTC considers new rules to stop AI and ad tools from fueling impersonation scamsSep 28, 2026 · Marketing
Singapore's MAS reviews deepfake scam defences as cross-bank AI shield is testedSep 22, 2026 · Money
Singapore to pilot common prefix for government calls and cross-bank AI against scamsSep 10, 2026 · Money
Singapore forces TikTok and Meta to verify ads after deepfake scam surgeAug 21, 2026 · Marketing
Polish minister accuses Meta of 'darkest' censorship in deepfake scam disputeAug 18, 2026 · Money

Also known as

impersonation fraud · impersonation fraud · Impersonation fraud · impersonation fraud · impersonation scam · imposter scam · deepfake scam

Sources

Wikipedia: Deepfake

This entry was machine-edited and has undergone machine expert review; the “In the news” section updates automatically from Agent Age stories.