Global Online Scam Trends in 2026: What to Expect in 2027

Global online scam trends in 2026: a wolf disguised as a financial broker persuades an older person to approve an online payment

Global Online Scam Trends in 2026

Online scams are becoming more international, more personalised and more difficult to judge by appearance alone.

During the first half of 2026, 69% of the organisations participating in Consumers International’s Scams Barometer said scam activity had increased in their jurisdictions. More than half identified artificial intelligence as the leading challenge. In a separate 2026 operation across 97 countries and territories, INTERPOL identified more than 142,000 victims and intercepted USD 293 million linked to social-engineering fraud.

There is not yet a reliable total for worldwide scam losses during 2026. There is also no authoritative projection for total worldwide losses in 2027. Available evidence nevertheless points to continued growth in impersonation, AI-assisted deception, hybrid investment and relationship scams, and internationally organised scam infrastructure.

The practical consequence is simple: professional writing, an attractive website, a familiar logo or even a convincing voice or video call can no longer establish that an offer is genuine. What matters is whether the claims can be supported by independent public evidence.

What the 2026 international evidence shows

1. Scam activity is still increasing across regions

Consumers International’s H1 2026 Scams Barometer collected information from 23 participants across 17 countries and five regions. Sixty-nine percent reported increased scam activity, while 52% identified AI as the leading threat, particularly through deepfakes, voice cloning, impersonation and deceptive advertisements or listings.

This was a survey of organisations in the anti-scam ecosystem, not a population survey of all consumers. It is therefore an indicator of direction rather than a measurement of the percentage of people being scammed.

Source: Consumers International, H1 2026 Scams Barometer

2. Social engineering is operating at international scale

INTERPOL’s Operation First Light ran from 15 January to 30 April 2026. The operation involved 97 countries and territories, resulted in 5,811 arrests, intercepted USD 293 million in illicit assets and identified more than 142,000 victims.

The operation targeted business email compromise, sextortion, romance scams, impersonation, investment scams and the associated movement of money. These figures describe one coordinated enforcement operation. They are not a worldwide total for 2026.

Source: INTERPOL, Operation First Light 2026

3. Fraud is now treated as a systemic financial threat

The Financial Action Task Force reported in February 2026 that 156 jurisdictions, representing 90% of the jurisdictions it had assessed, identified fraud as a major money-laundering risk.

This matters because online scams are not isolated messages sent by individuals. The proceeds can move through bank accounts, payment services, money mules, cryptocurrency services and professional laundering networks spanning several countries.

Source: FATF, Cyber-Enabled Fraud

4. The most common approaches still begin with communication

The OECD’s Consumer Finance Risk Monitor 2026 found that responding authorities most frequently identified:

  • phishing, voice phishing or text-message phishing for personal information: 83%;
  • criminals posing as banks, advisers or financial-services providers: 78%; and
  • fake payment, investment or insurance opportunities: 67%.

These percentages describe the views of responding authorities about the types affecting consumers. They are not percentages of the world population.

Source: OECD, Consumer Finance Risk Monitor 2026

5. AI improves scale and credibility, not just grammar

INTERPOL’s 2026 Global Financial Fraud Threat Assessment describes financial fraud as one of the most serious and rapidly evolving transnational crimes. It reports that AI-enhanced fraud can be 4.5 times more profitable than traditional methods.

AI can help an operator translate messages, adapt wording to a victim’s circumstances, produce images and documents, clone voices and continue many conversations at once. It can also help combine several familiar approaches into one operation.

Source: INTERPOL, Global Financial Fraud Threat Assessment 2026

Why there is no trustworthy global 2026 loss total yet

Worldwide scam statistics are assembled from national police reports, consumer complaints, banks, payment networks, regulators, blockchain analysis and surveys. These sources use different definitions and reporting periods.

Many incidents are never reported. One source may count attempted scams, another confirmed payments, and another suspicious transactions. A report published in 2026 may also describe losses that occurred during 2025.

For that reason, a figure should not be called a “2026 scam total” merely because the report containing it was published in 2026. AdUncovered will update this article when comparable full-year information becomes available.

What is reasonably expected – scams 2027

The available forecasts support a direction of travel, but not a precise worldwide loss figure.

More impersonation across text, voice and video

Deepfake audio and video are expected to become increasingly ordinary components of impersonation. They can be used alongside compromised accounts, copied branding and information collected from public profiles.

Gartner forecasts that AI agents will reduce the time required to exploit exposed accounts by 50% by 2027. That prediction concerns account takeover rather than every type of consumer scam, but it illustrates how automation can shorten the period between acquiring information and using it.

Source: Gartner, account-exposure forecast for 2027

Faster and more personalised targeting

Messages can increasingly be adapted to a person’s occupation, interests, language, location or recent activity. A suspicious approach may no longer look generic. The same operation can test different narratives and continue with whichever one receives a response.

Hybrid scams rather than neat categories

The labels “romance scam”, “investment scam” and “impersonation scam” can suggest separate events. In practice, the same approach may begin with a relationship or professional opportunity, move to an investment platform, introduce cryptocurrency or fees, and later use threats or impersonated recovery services.

INTERPOL’s assessment specifically identifies the increasing integration of sextortion into romance and investment fraud and the sharing of technology and laundering services between criminal networks.

A wider geographic spread of scam infrastructure

INTERPOL has identified scam centres beyond the locations with which they were previously associated. Its global assessment expects significant escalation in fraud during the next three to five years as technology, criminal collaboration and fraud-as-a-service reduce the cost of operating internationally.

Continued pressure on banks and payment providers

Deloitte has forecast that generative-AI-enabled fraud losses affecting US financial institutions and their customers could reach USD 40 billion in 2027, compared with USD 12.3 billion in 2023.

That is a US banking forecast, not a projection for worldwide scam losses. It should not be presented as a global figure. It is useful only as an indication of the expected scale of AI-related pressure in one large financial market.

Source: Deloitte, deepfake banking and AI fraud risk

What these trends change for an ordinary customer

The traditional advice to look for bad grammar or a poorly designed website is no longer sufficient. Neither is a single positive review or an apparently genuine social-media profile.

The useful question is not only, “Does this look convincing?” It is:

What independent evidence supports the important claims being made?

That means separating several different kinds of evidence.

EvidenceWhat it may establishWhat it does not establish by itself
A functioning websiteThe site was accessible when checkedWho operates it or whether its claims are true
HTTPS certificateThe connection was encrypted and the certificate was validThe business is trustworthy
Domain-registration informationAvailable dates, status and nameserver informationThe identity or intentions of the person contacting you
Trustpilot complaintA reviewer publicly described an experienceThat every allegation is correct or applies to your case
ScamAdviser ratingAn automated service detected and weighted certain signalsA verified finding that a site is safe or fraudulent
Company Terms/About/Contact pagesWhat the site says about itselfIndependent confirmation of those statements
Regulator warningWhat an identified authority publishedThat an unlisted offer is automatically safe
No public complaints foundNo relevant result was returned from the sources checkedThat no complaint exists or that the offer is legitimate

A practical evidence check before acting

Before paying, creating an account, uploading identity documents, connecting a wallet or installing software, preserve the offer and examine:

  1. The claim: What exactly is being promised, and what must you do or pay?
  2. The identity: Is the person or business independently connected with the claimed organisation?
  3. The destination: Where does the submitted link redirect, and what is the final domain?
  4. The site’s own statements: Do its Terms, About, Contact and payment information agree with the approach you received?
  5. The public pattern: Are there closely matching complaints, warnings or reports describing the same sequence?
  6. The technical history: What public domain, certificate, DNS and archive information is available?
  7. The contradictions: Which important claims conflict with another source?
  8. The unknowns: Which important questions could not be answered from available public information?

A lack of evidence is not proof of wrongdoing. It is also not proof of legitimacy. It is an uncertainty that should be shown as such.

How an AdUncovered Quick Report helps

A Quick Report is a focused check of one seller and the advertised offer before you buy. We use available public information to answer your questions and highlight material concerns within the Quick scope.

The research covers serious warning signs, scam reports, customer complaints and reviews, and comparison of the advert with available published offer details, prices and important conditions. We check the seller’s identity sufficiently to help match the evidence to the correct business.

Ordinary research covers the past year. Relevant court orders and trading restrictions have no age cutoff; where their current status cannot be confirmed, the report says so.

Findings are not limited to a fixed number. Repeated findings are combined with their sources, complaints are distinguished from established facts, and conflicting or unavailable information is explained. The downloadable PDF includes source references, research dates and limitations.

Quick does not investigate your individual transaction or prepare a refund claim. Choose Comprehensive for broader research, normally covering the past three years, including relevant transaction details and refund-related information you supply.

Provide a usable public link identifying the seller, shop, profile, listing or offer, together with relevant written details. A business does not need its own website.

Neither report guarantees safety, determines criminal guilt, provides legal advice or recovers money. No negative findings does not establish that an offer is safe: public records may be incomplete, publication may be delayed and new information may appear later.

Frequently asked questions

Are online scams increasing in 2026?

The available international indicators point to continued growth. In Consumers International’s H1 2026 barometer, 69% of participating organisations reported increasing scam activity. This is an ecosystem survey rather than a worldwide victim count.

How much money will be lost to scams in 2027?

There is no authoritative projection for total worldwide scam losses in 2027. Forecasts that cover one country, payment channel or fraud category should not be presented as a global number.

Can AI make a fake offer look professional?

Yes. AI can improve writing, translation, images, documents, voice imitation and personalised messaging. Professional presentation is therefore not independent evidence that an offer is genuine.

Does an HTTPS certificate mean a website is legitimate?

No. A valid certificate protects the connection between the browser and the website. It does not verify the site’s business claims or the identity of the person who sent the link.

Can an AdUncovered report confirm that an offer is safe?

No. It presents source-linked public findings, distinguishes them from customer-supplied information and explains what could not be confirmed. It does not make a legitimacy determination or safety guarantee.

Sources and methodology

This article uses international organisations, intergovernmental bodies and clearly scoped industry forecasts. Statistics are labelled according to what they actually measure. Reports published in 2026 but measuring 2025 activity are not represented as full-year 2026 loss statistics.

This article was researched with automation and reviewed against the linked source material. Sources were last checked on 25 September 2026.

Check before you pay

About to pay a seller or business you don’t know? Start with a Free Quick Check of the advert link. It screens for common warning signs and shows where each result came from.

If you need the seller researched, a Quick Report or Comprehensive Report brings together source-linked Search Results on the seller and the offer. Neither gives a verdict: you check the sources and decide for yourself.

Similar Posts