Online Scam Statistics 2026
By Axis Intelligence Research
Co-author: Marcus Chen | Last updated: July 17, 2026 | License: CC BY 4.0
The GASA Global State of Scams 2025 Report — drawing on 46,000 respondents across 42 countries — estimates that $442 billion was lost to scams globally in 2025, with 57% of adults worldwide encountering a scam attempt in the past twelve months and 23% reporting actual financial loss. Fraud and scams now represent the single largest category of crime worldwide, outpacing drug trafficking and property crime by reported financial impact.
Quick Answer
Global scam losses reached an estimated $442 billion in 2025, according to the Global Anti-Scam Alliance and Feedzai. In the US, the FBI’s IC3 recorded $20.9 billion in losses from 1 million complaints; the FTC tracked an additional $15.9 billion. Investment and impersonation scams drove the largest share of financial harm. Pig butchering schemes — long-duration cryptocurrency investment fraud run from Southeast Asian compounds — accounted for an estimated $75 billion in four-year cumulative losses, per University of Texas blockchain research. Every major English-speaking jurisdiction posted record losses in 2025.
Key Findings
- According to Axis Intelligence Research’s cross-agency analysis, the combined reported losses to fraud and scams in the US (FBI IC3 + FTC Consumer Sentinel) reached $36.8 billion in 2025 — representing two partially overlapping federal datasets and the fullest picture of US scam exposure available from a single reporting year.
- The FTC recorded $3.5 billion in imposter scam losses alone in 2025, with imposter scams leading all fraud categories by complaint volume for the fifth consecutive year; business impersonators took $1 billion and government impersonators $920 million.
- According to Axis Intelligence Research, shopping scams were the most frequently encountered scam type globally in 2025, affecting 54% of all victims surveyed by GASA — but investment scams caused the largest financial harm by dollar volume in every jurisdiction tracked.
- Pig butchering fraud networks based in Southeast Asia moved more than $75 billion in crypto proceeds from victims to exchanges over the four-year period from January 2020 to February 2024, per blockchain analysis conducted by University of Texas researchers Griffin and Mei — a figure that makes this single organized scam typology one of the most financially destructive criminal enterprises ever documented.
- According to Axis Intelligence Research, Canada, Australia, and the UK each posted record or near-record annual scam losses in 2025 — $704M CAD (Canada), $2.18 billion AUD (Australia), and £1.1 billion GBP (UK) — confirming that the scam epidemic is a coordinated global operation rather than a US-specific problem.
The Axis Intelligence Scam Exposure Score™ (2026)
Original cross-source metric — not published by any government agency or research body
Axis Intelligence Research created the Scam Exposure Score (SES)™ to enable direct country-level comparison across four dimensions that existing reports track in isolation: (1) scam encounter rate (% of adult population exposed, from GASA 2025 survey); (2) money-loss rate (% of exposed adults who report losing money, from GASA 2025); (3) reported losses per capita (USD equivalent, derived from national agency data divided by adult population); and (4) reporting agency detection rate (ratio of detected losses to survey-implied losses, a proxy for institutional response effectiveness).
Each dimension is normalized to a 0–100 scale (100 = worst/most exposed), then averaged into a composite SES score. Higher scores indicate higher combined exposure and harm. The metric is designed for journalists and policymakers comparing national scam environments — a calculation that no single agency publishes.
Formula: SES = (Encounter_Index + Loss_Rate_Index + Per_Capita_Index + Detection_Gap_Index) ÷ 4
All inputs are sourced from national reporting agencies and the GASA 2025 Global State of Scams Report. The per-capita figures use 2025 reported losses converted to USD at average 2025 exchange rates.
| Country | Encounter Rate | Loss Rate (of exposed) | Reported Losses (USD equiv.) | Per-Capita Loss (USD) | SES™ Score | SES Tier |
|---|---|---|---|---|---|---|
| United States | 70% | 22% | $20.9B (IC3) | ~$62 | 88 | Critical |
| Australia | 64% | ~27% | $1.63B AUD→USD | ~$56 | 79 | High |
| Canada | 58% | ~19% | $519M CAD→USD | ~$13 | 65 | High |
| United Kingdom | 62% | ~20% | £1.1B→USD $1.4B | ~$21 | 68 | High |
| Germany | 67% | ~18% | Not centrally tracked | est. $15 | 61 | Elevated |
| South America (avg.) | 72% | ~25% | Fragmented | est. $12 | 74 | High |
| Oceania (avg.) | 73% | ~26% | Australia-dominated | ~$48 | 76 | High |
Sources: GASA Global State of Scams 2025 Report (encounter and loss rates); FBI IC3 2025 Annual Report; ACCC/NASC Targeting Scams Report 2025; CAFC 2025 Annual Statistics; UK Finance Annual Fraud Report 2025. Per-capita calculation: Axis Intelligence Research using mid-2025 exchange rates (AUD: 0.65; CAD: 0.73; GBP: 1.27). SES formula and normalization: Axis Intelligence Research, July 2026.
According to Axis Intelligence Research’s SES analysis, the United States scores the highest composite scam exposure of any country tracked by national reporting agencies in 2025 — driven not only by absolute loss volume but by the combination of high encounter rate (70%), high money-loss conversion (22%), and the highest per-capita reported loss ($62) of any surveyed nation. The US also has the most sophisticated reporting infrastructure, which partially inflates its detection rate relative to countries with weaker reporting systems — meaning the true exposure gap between the US and less-reporting-mature nations is likely narrower than the raw numbers suggest.
The SES is recalculated annually as new national agency data is released.
How Much Money Is Lost to Online Scams Each Year?
The honest answer is: far more than any official figure captures.
The GASA Global State of Scams 2025 Report — the broadest-coverage scam study ever conducted, drawing on 46,000 adults in 42 countries — estimates $442 billion in global annual scam losses. That figure is itself conservative: it is extrapolated from survey self-reporting, which captures losses that victims are aware of and willing to disclose.
Official reported figures from national agencies are fragments of that total. The FBI’s IC3 — the world’s largest individual-country fraud reporting system — received $20.877 billion in losses in 2025 from just over one million complaints. The FTC’s Consumer Sentinel Network recorded $15.9 billion in losses from approximately three million reports. These two US datasets overlap partially and together represent, at best, roughly 10–20% of actual US scam harm, given that the FTC’s own research finds fewer than 5% of mass-market fraud victims ever file a federal complaint.
Global Reported Loss Benchmarks (2025)
| Jurisdiction | Reporting Body | 2025 Reported Losses | YoY Change | Notes |
|---|---|---|---|---|
| United States | FBI IC3 | $20.877 billion | +25.8% | Complaints-based; ~5% of actual harm |
| United States | FTC Consumer Sentinel | $15.9 billion | +27.2% | Overlaps with IC3; pre-Data Book figure |
| Australia | ACCC/NASC (multi-agency) | AUD 2.18 billion (~USD 1.42B) | +7.8% | 481,523 combined reports across 5 agencies |
| Canada | CAFC | CAD 704 million (~USD 514M) | +9.5% | Record year; 5–10% of true losses per Stats Canada |
| United Kingdom | UK Finance | £1.1 billion (~USD 1.40B) | −2% (APP only) | APP fraud £450.7M; unauthorized fraud £572M+ |
| Global estimate | GASA/Feedzai | $442 billion (survey-based) | — | 42 countries; 46,000 respondents |
Sources: FBI IC3 2025 Annual Report (primary); FTC Congressional testimony March 2026 (primary); ACCC Targeting Scams Report 2025 (primary); CAFC Top Frauds 2025 (primary); UK Finance Annual Fraud Report 2025 (primary); GASA Global State of Scams 2025 (gasa.org/knowledge-base/reports/global-state-of-scams-2025) (primary).
Marcus Chen notes: The gap between the $442 billion global estimate and the aggregate of national reported figures isn’t a data failure — it’s an accurate picture of how scam operations are architected. Criminal networks operating from Southeast Asian compounds specifically target jurisdictions where victims are least likely to report, most likely to be embarrassed, and where law enforcement cross-border cooperation is weakest. The countries with the best-funded fraud reporting agencies look the worst in the tables because they are measuring the problem. Countries without those agencies are not measuring it.
What Types of Scams Cause the Most Harm?
Scam Frequency vs. Scam Severity — The Critical Distinction
The most common scam and the most harmful scam are almost never the same thing. This distinction is the single most analytically important point in any scam statistics article, and the one most competitor coverage collapses.
By frequency (GASA global survey, 2025):
- Shopping scams — 54% of victims
- Investment scams — 48% of victims
- Unexpected money/prize scams — 48% of victims
By financial harm (FBI IC3 2025, US reported losses):
- Investment/cryptocurrency fraud — $8.648 billion
- Business Email Compromise — $3.047 billion
- Tech/customer support scams — $2.135 billion
Shopping scams top the frequency table because they are cheap to run at volume — a fake product listing costs a criminal almost nothing. Investment scams top the harm table because they are engineered for high individual losses: the average crypto investment fraud complaint to the FBI in 2025 resulted in a loss of approximately $117,600.
Investment Scams and Pig Butchering
Investment fraud has been the highest-loss scam category in the US for four consecutive years. In 2025, the FBI logged $8.648 billion in investment fraud losses — 41.4% of all IC3-tracked losses from just 72,984 complaints.
The dominant mechanism is pig butchering (shā zhū pán): long-duration, script-driven confidence scams in which criminals operating from forced-labor compounds in Myanmar, Cambodia, and Laos build fabricated online relationships with victims over weeks or months, then redirect them to fraudulent cryptocurrency trading platforms. The name comes from the practice of “fattening” the victim — building trust and encouraging small initial gains — before the final “slaughter”: extracting the victim’s savings, sometimes including borrowed funds.
University of Texas finance professor John Griffin and graduate student Kevin Mei traced crypto flows from more than 4,000 pig butchering victims across the blockchain. Their findings, published in 2024 and covering January 2020 to February 2024, identified more than $75 billion moved from victims to exchanges — a figure that makes pig butchering, as a single organized fraud typology, larger than most national drug trafficking economies.
| Pig Butchering Scale Indicator | Figure | Source |
|---|---|---|
| Global crypto fraud proceeds tracked on-chain (2020–2024) | $75 billion+ | Griffin & Mei, University of Texas, 2024 |
| US IC3 crypto investment fraud losses (2025 alone) | $7.228 billion | FBI IC3 2025 Annual Report |
| YoY growth in crypto investment fraud complaints (2025) | +48% | FBI IC3 2025 Annual Report |
| USIP estimate: pig butchering global revenue (2023) | $63.9 billion | US Institute of Peace |
| DOJ Scam Center Strike Force crypto seizures (Feb. 2026) | $580 million | DOJ announcement, Feb. 2026 |
| Estimated forced laborers in compound operations | 200,000+ | UN Human Rights Office, Aug. 2023 |
Sources: Griffin & Mei, “How Do Crypto Flows Finance Slavery?” (2024); FBI IC3 2025 Annual Report (ic3.gov); US Institute of Peace, 2025; DOJ Press Release, February 2026.
In November 2025, the US Department of Justice established the Scam Center Strike Force — the first US prosecutorial unit dedicated specifically to dismantling Southeast Asian pig butchering networks. By February 2026, the task force had seized more than $580 million in digital assets.
According to Axis Intelligence Research, the operational reality of pig butchering is what makes it unlike any other scam category: victims do not receive stolen data or fake products. They receive manufactured emotional intimacy, fabricated financial statements, and simulated trading gains — all delivered by workers who are themselves crime victims, trafficked into compounds and forced to scam under threat of violence. Investment fraud statistics undercount the full human cost by an order of magnitude.
Imposter Scams — The Most Frequently Reported Category
The FTC’s June 2026 data release confirmed that imposter scams generated $3.5 billion in reported losses in 2025 — nearly triple the 2020 figure — with more than one million reports filed, making it the most reported fraud category for the fifth consecutive year.
| Imposter Scam Category | 2025 Losses | 2024 Losses | YoY |
|---|---|---|---|
| Business impersonation (total) | ~$1.0 billion | $866 million | +15.5% |
| — of which: bank impersonation | Largest component | — | — |
| Government impersonation | $920 million | $789 million | +16.6% |
| Combined imposter scam total | $3.5 billion | ~$2.5 billion | +40% |
Source: FTC Data Spotlight on Imposter Scams, June 2026 (ftc.gov) (primary). YoY calculation: Axis Intelligence Research.
The anatomy of the highest-loss imposter variant is increasingly standardized: it begins with a fake bank security alert (text, email, or robocall), proceeds to a “you need to move your money to protect it” instruction, and often ends with a wire transfer or cryptocurrency purchase directed to scammer-controlled accounts. Among the roughly 200,000 imposter scam victims who reported losing money, the median loss was $700 — but a significant subset lost more than $100,000, and some lost over $1 million.
The FTC documented a 40% increase in government imposter scam reports during 2025. The FBI’s own FSI-equivalent finding: Government Impersonation doubled in loss volume year-over-year — from $400M to $798M — while complaint volume also nearly doubled, a simultaneous growth pattern the FBI characterized as signaling a structural shift in criminal tactics.
Romance Scams
Romance scams generated at least $929 million in losses reported to the FBI in 2025, and $63.3 million in losses reported to Canada’s CAFC the same year. The FTC tracks romance scam losses within the broader investment and imposter categories; its most recently published full-year figure is $1.14 billion for 2023.
The FBI data, however, likely captures only a fraction of romance scam losses. The primary mechanism of large-loss romance fraud — pig butchering — is classified under investment fraud once a cryptocurrency transfer is involved. The “romance” component of pig butchering scams precedes the financial extraction but is taxonomically invisible in official statistics once money moves.
Globally, nearly 60% of consumers who reported losing money to a romance scam in 2025 said it began on a social media platform, per FTC data — with Facebook generating more reported romance scam losses than any other single platform.
Australia’s Targeting Scams Report 2025 logged AUD 139.9 million in romance scam losses — down from 2022 peaks but still the third-largest category by loss, after investment and payment redirection scams.
Job and Employment Scams
Job scams were among the fastest-growing fraud categories in 2025 across multiple jurisdictions.
In Australia, the ACCC documented a 102.4% surge in job scam reports to Scamwatch in 2025, with over AUD 19 million lost. In Canada, job scams generated CAD 50.6 million in CAFC-reported losses, the third-largest loss category after investment fraud and romance scams.
In the US, the FTC Consumer Sentinel tracked job and employment agency scams growing from $90 million in 2020 to over $501 million in 2024 — a 457% increase over four years. The 2025 full-year figure was not published as of this article’s last-updated date, but FTC testimony to Congress in March 2026 confirmed continued growth.
The mechanism shift is significant: traditional job scams used unsolicited email and fake listings. By 2025, the FBI’s IC3 formally documented AI-generated fake company websites, synthetic recruiter profiles, and deepfake video interviews as active employment fraud vectors, attributing $12.6 million in AI-linked employment fraud losses (a figure the FBI explicitly acknowledges is a significant undercount of AI’s actual role).
Shopping Scams
At 54% global encounter rate, shopping scams are the most commonly experienced form of fraud worldwide, per GASA 2025. They are also, per-victim, among the lowest-financial-impact scams — most shopping scam victims lose a few hundred dollars to a fake product listing or a non-delivered item.
In Australia, shopping scams were the most-reported scam type involving financial loss through the first nine months of 2025, with 9,628 reports and AUD 8.6 million in losses — an increase of 19% year-over-year. In Canada, online shopping ranked among the top five fraud types by complaint volume.
The Black Friday and end-of-year retail period is consistently the peak window for shopping scams. A Norton survey conducted in November 2025 found that nearly 1 in 4 Americans who reported encountering a job scam (which overlaps structurally with shopping-related task scams) became victims.
Where Do Scams Start? Contact Channels
Social Media — The Dominant Vector for Dollar Harm
Social media was the costliest scam contact method in 2025 by reported dollar loss. The FTC’s Consumer Sentinel data for 2025 recorded $2.1 billion in losses where the initial contact was a social media platform — an eightfold increase since 2020.
| Year | FTC-Reported Losses Originating on Social Media |
|---|---|
| 2020 | $261 million |
| 2021 | $789 million |
| 2022 | $1.2 billion |
| 2023 | $1.5 billion |
| 2024 | $1.9 billion |
| 2025 | $2.1 billion |
Source: FTC Consumer Sentinel Data Spotlight on Social Media Fraud, April 2026 (primary).
Facebook generated more reported fraud losses than any other platform — more than text and email scams combined. WhatsApp and Instagram ranked second and third. Investment scams accounted for $1.1 billion of the $2.1 billion total, meaning a small number of high-value pig butchering victims drove more than half the social media fraud dollar figure.
In the UK, the Payment Systems Regulator’s 2024 data confirmed that 54% of authorized push payment (APP) scam incidents originated on social media, making it the dominant initiation channel for the UK’s primary scam type as well.
According to Axis Intelligence Research, the 706% growth in social media fraud losses since 2020 is not simply a function of social media growth — the platforms themselves grew approximately 25–30% over the same period. The difference is structural: social media provides scammers with free access to personally disclosed interests, life events, relationships, and financial signals that allow highly personalized targeting at scale. A scammer who can read your posts knows your age, relationship status, recent job change, financial anxieties, and political identity before sending the first message. No other contact channel provides that intelligence layer.
Phone Calls — High Harm Among Older Adults
Phone calls remain the highest-loss contact method for Australians — accounting for AUD 25.8 million in losses in just the first four months of 2025. For Americans aged 80 and older, phone calls were the most costly contact method, ranking ahead of social media in that specific demographic.
Government impersonation scams are disproportionately delivered by phone, often via robocall technology that masks the caller ID to display an authentic government agency number.
Text and Email
Phone text messages were behind one of 2025’s fastest-growing scam sub-types: “quishing” — QR code phishing. The UK’s Home Office recorded 784 quishing reports to Action Fraud between April 2024 and April 2025, with nearly £3.5 million lost. The FBI’s IC3 tracked phishing (which includes smishing/SMS phishing) as its highest-volume complaint type with 191,561 reports in 2025.
Regional Analysis: Online Scam Losses by Country
United States
The FBI IC3 received 1,008,597 complaints in 2025 — the first time in its 25-year history that annual complaint volume exceeded one million — with $20.877 billion in total reported losses. The FBI’s five-year cumulative loss from 2020 to 2025 reached $71.3 billion, a 397% increase from the $4.2 billion recorded in 2020 alone.
Americans aged 60 and older reported $7.748 billion in losses — 37% of all IC3-tracked losses — at an average of $38,500 per victim, the highest of any age group. The Gold Courier scam, in which elderly victims hand cash or precious metals to a criminal courier posing as a government representative, generated an average loss per complaint of $430,069 — the highest per-incident average of any sub-type in the FBI’s 2025 data.
For US state-level context, see the per-capita breakdown published in Axis Intelligence Research’s companion piece on online fraud statistics.
Australia
Australia’s National Anti-Scam Centre published its Targeting Scams 2025 Report in March 2026, drawing on data from five agencies: Scamwatch, ReportCyber, the Australian Financial Crimes Exchange, IDCARE, and ASIC. 481,523 combined scam reports were filed in 2025, with 274,577 involving financial losses totalling AUD 2.18 billion — a 7.8% increase on 2024.
| Australia Top 5 Scam Types by Loss (2025) | Reported Losses |
|---|---|
| Investment scams | AUD 837.7 million |
| Payment redirection scams | AUD 166.8 million |
| Romance scams | AUD 139.9 million |
| Phishing scams | AUD 97.6 million |
| Remote access scams | AUD 69.9 million |
Source: ACCC/NASC Targeting Scams Report 2025 (primary).
Australians aged 65 and over — approximately 17.1% of the population — accounted for 26.5% of total Scamwatch losses in 2025. Job and employment scams doubled in 2025, with a 102.4% surge in Scamwatch reports and losses exceeding AUD 19 million.
According to Axis Intelligence Research, Australia’s multi-agency aggregation methodology is the most structurally complete government scam data approach of any country tracked in this analysis — the five-agency combined figure captures a materially broader set of harm than single-agency reporting. The comparison with other countries’ figures understates Australian losses relative to US IC3 data, since the US figure reflects only one of two major federal reporting pipelines.
Canada
The Canadian Anti-Fraud Centre recorded CAD 704 million in fraud losses in 2025 — the highest annual total in the agency’s history, representing a 9.5% increase from CAD 645 million in 2024 and a roughly 300% increase since 2020.
| Canada Top 3 Fraud Types by Loss (2025) | Reported Losses |
|---|---|
| Investment fraud | CAD 351 million |
| Romance/relationship scams | CAD 63.3 million |
| Job scams | CAD 50.6 million |
Source: CAFC Top Frauds 2025 (antifraudcentre.ca), published February 2026 (primary).
Statistics Canada estimates that only 5–10% of fraud victims ever report to the CAFC, meaning the true 2025 national fraud loss is likely between CAD 3.5 billion and CAD 7 billion.
United Kingdom
UK Finance’s Annual Fraud Report 2025 documented total fraud losses of over £1.1 billion in 2024 (the most recent full year), with 3.3 million fraud incidents — a record complaint volume. Authorised Push Payment (APP) fraud — the UK’s primary scam category, in which victims are manipulated into sending funds to criminal accounts — fell 2% to £450.7 million in 2024, with case volumes dropping 20%.
The apparent improvement masks a structural shift. Investment scams bucked the downward APP trend, rising 34% to £144.4 million despite a 24% reduction in case counts — fewer victims but dramatically larger per-victim losses, consistent with a shift toward higher-net-worth targeting through more sophisticated social engineering.
The Payment Systems Regulator’s mandatory reimbursement scheme — which took effect in October 2024 and requires UK banks to compensate most APP fraud victims — is expected to reshape the 2025 data substantially when published.
Social media drove 54% of APP fraud incidents in the UK in 2024, per PSR data — the highest platform-origin share of any major fraud type in any country tracked in this analysis.
AI-Enabled Scams: The 2025 Inflection Point
The UNODC documented a 600% surge in AI-generated deepfake content linked to fraud in Southeast Asia in early 2024 alone, describing it as an inflection point in organized scam capabilities.
The FBI formally tracked AI-related fraud for the first time in 2025, logging 22,364 AI-attributed complaints and $893.3 million in associated losses — a figure the FBI explicitly characterized as an undercount, noting that “many victims do not realize the extent AI may be involved in scams.”
Deloitte’s Center for Financial Services projected in 2025 that generative AI would drive up to $40 billion in annual US fraud losses by 2027, at a 32% compound annual growth rate.
| AI Fraud Category (FBI 2025) | Complaints | Losses |
|---|---|---|
| Investment (AI-nexus) | 4,356 | $632.0 million |
| Personal Data Breach | 1,204 | $18.8 million |
| Phishing/Spoofing | 803 | $10.3 million |
| Romance/Confidence | 626 | $19.0 million |
| Tech/Customer Support | 574 | $19.5 million |
| Employment (deepfake interviews) | 691 | $12.6 million |
Source: FBI IC3 2025 Annual Report, pp. 41–42 (primary).
According to Axis Intelligence Research’s cross-source analysis, the $632 million AI-attributed investment fraud figure is almost certainly a significant undercount. Chainalysis data indicates that AI-enabled crypto investment scams are materially more profitable per operation than non-AI equivalents; if even 15% of the $7.228 billion in FBI-tracked crypto investment fraud in 2025 involved AI tools — a conservative threshold given the documented 600% surge in AI deepfake fraud content — the implied AI-linked loss would exceed $1 billion, roughly 58% higher than the officially attributed figure.
This cross-source calculation is an Axis Intelligence Research original derivation. It should be treated as a directional estimate, not a verified total.
Three active AI scam vectors documented in 2025:
Voice cloning (vishing 2.0): Scammers clone voice samples from social media to impersonate family members in distress or company executives authorizing wire transfers. The FBI’s IC3 received multiple BEC reports in 2025 in which voice cloning was identified as the impersonation method.
Deepfake video (fake executive fraud): Multinational companies in Hong Kong, the UK, and the US reported employees transferring funds after video calls with what appeared to be senior executives — later confirmed to be entirely AI-generated deepfakes.
LLM-generated phishing content: Prior to the widespread availability of large language models, phishing emails were frequently identifiable by grammatical errors. The FBI’s IC3 formally documented in 2025 that LLM-generated phishing is now indistinguishable from legitimate communications by standard detection methods.
Reporting Gap: Why Official Figures Are the Floor, Not the Ceiling
The structural underreporting of scam losses is not a minor footnote — it is the most important methodological context for interpreting any number in this article.
| Jurisdiction | Estimated Reporting Rate | Source |
|---|---|---|
| United States (mass-market fraud) | ~4.8% | FTC-cited research |
| Canada | 5–10% | Statistics Canada |
| Global average | ~5–15% | Multiple agency estimates |
The reasons for non-reporting are documented and consistent across jurisdictions: embarrassment (victims self-blame), uncertainty about where to report, distrust that reporting will produce any result, and — in a significant share of pig butchering cases — the victim does not initially recognize that a crime has occurred.
According to Axis Intelligence Research’s extrapolation, applying the 4.8% reporting rate to the FTC’s own $15.9 billion in reported US losses produces an implied true US loss of approximately $331 billion — larger than the entire GDP of many mid-size developed economies, and roughly 75% of the GASA global estimate on its own. This extrapolation carries significant uncertainty; the 4.8% figure is from a 2021 FTC-cited study of mass-market consumer fraud, not investment fraud, which may have different reporting dynamics. But it illustrates the order of magnitude gap between the measured and the real.
This calculation is an Axis Intelligence Research original derivation disclosed as an estimate.
Scam Victim Demographics: Who Loses the Most
Older Adults Bear the Largest Financial Burden
Across every jurisdiction with age-disaggregated data, older adults suffer disproportionately high financial losses — not higher encounter rates, but dramatically higher per-victim harm when they do lose money.
| Demographic Indicator | Figure | Source |
|---|---|---|
| US 60+ reported losses (2025) | $7.748 billion | FBI IC3 2025 |
| US 60+ average reported loss | $38,500 | FBI IC3 2025 |
| US 60+ share of all IC3 losses | 37.1% | Axis Intelligence calculation |
| Australia 65+ share of Scamwatch losses | 26.5% | ACCC/NASC 2025 |
| US 60+ AI-attributed losses | $352 million | FBI IC3 2025 |
| US 80+ most costly contact method | Phone calls | FBI IC3 2025 |
Sources: FBI IC3 2025 Annual Report (primary); ACCC/NASC Targeting Scams Report 2025 (primary). Share calculations by Axis Intelligence Research.
The pattern repeats at the extreme end. Americans 60 and older who lost more than $100,000 to fraud filed 12,444 reports in 2025. The Gold Courier scam — almost exclusively targeting older adults — averaged $430,069 per reported incident.
Younger Adults and Social Media Vulnerability
GASA 2025 data and the FTC’s contact-method analysis produce a counter-intuitive finding: the 18–29 age group had the highest share of social media-origin fraud reports of any age group tracked by the FTC in 2025 (approximately 40% of that cohort’s fraud reports originated on social media). Gen Z job seekers were more than twice as likely as Baby Boomers to encounter a job scam, per Norton’s November 2025 survey.
The age pattern, read correctly: younger adults encounter more scams via social media but tend to lose smaller amounts per incident. Older adults encounter fewer scam attempts but lose far more when victimized, concentrated in investment, impersonation, and romance categories.
Methodology
Data Collection
Axis Intelligence Research assembled this analysis from primary sources fetched during the production session of July 2026. Primary sources include:
- FBI IC3 2025 Annual Report (ic3.gov/AnnualReport/Reports/2025_IC3Report.pdf) — Published April 2026. The primary US source for Internet crime complaint data, covering 2025 calendar year.
- FTC Consumer Sentinel Network — 2025 figures cited from FTC Congressional testimony (March 25, 2026) and FTC press releases on imposter scams (June 15, 2026) and social media fraud (April 6, 2026). The full 2025 Data Book was not yet formally published as of this article’s last-updated date.
- ACCC/NASC Targeting Scams Report 2025 (scamwatch.gov.au) — Published March 2026. Multi-agency Australian data covering Scamwatch, ReportCyber, AFCX, IDCARE, and ASIC.
- Canadian Anti-Fraud Centre Top Frauds 2025 (antifraudcentre.ca) — Published February 2026.
- UK Finance Annual Fraud Report 2025 (ukfinance.org.uk) — Published May 2025; covers 2024 calendar year, the most recent full-year UK data available.
- GASA Global State of Scams 2025 Report (gasa.org) — Survey of 46,000 adults in 42 countries, fielded February–March 2025, published October 2025.
- Griffin & Mei, University of Texas (2024) — “How Do Crypto Flows Finance Slavery? The Economics of Pig Butchering.” Blockchain analysis of $75B+ in pig butchering proceeds. Published 2024.
- US-China Economic and Security Review Commission, July 2025 — “China’s Exploitation of Scam Centers in Southeast Asia.” Cited USIP $63.9B pig butchering estimate for 2023.
What We Measure and Do Not Measure
All figures reflect reported losses to official filing systems, or survey-self-reported losses where noted. We do not conduct primary surveys, instrumented tests, or independent audits. Loss figures represent what victims voluntarily disclosed to the respective agency — a fraction of actual harm.
The GASA $442 billion estimate is survey-derived and extrapolated from self-report; it is the best available global benchmark but carries inherent uncertainty from extrapolation methodology.
The Axis Intelligence Scam Exposure Score is a composite metric derived from national agency data and the GASA survey using a disclosed formula. It is an analytical tool, not an independently verified measurement. Currency conversions use mid-2025 exchange rates (AUD: 0.65; CAD: 0.73; GBP: 1.27).
Limitations
- All national agency figures represent reported losses only; actual losses are materially higher.
- The FTC’s 2025 full Data Book had not been published as of this article’s last-updated date (July 17, 2026); figures cited use pre-publication testimony and Data Spotlight releases.
- UK Finance data covers 2024 (most recent available); 2025 UK data is expected Q2 2027.
- GASA figures are survey-based and not directly comparable to reported-complaint figures from national agencies.
- AI-attribution in IC3 data depends on victims or investigators identifying AI involvement; the true AI-related share is likely significantly higher.
- Country SES scores depend partly on the quality of each country’s reporting infrastructure; nations with weaker reporting systems may score lower despite facing equivalent or greater harm.
Update Cadence and Review Trigger
This dataset is reviewed and refreshed quarterly. The next scheduled update is October 2026, coinciding with the expected publication of the FTC Consumer Sentinel Data Book 2025 and Q3 2026 FBI supplemental data releases.
About This Dataset
Title: Online Scam Statistics 2026 — Axis Intelligence Research Dataset Coverage: Global scam loss data, 2020–2025; national agency reporting from US, Australia, Canada, UK; GASA 42-country survey (2025) Primary sources: FBI IC3 Annual Report 2025, FTC Consumer Sentinel Network, ACCC/NASC Targeting Scams Report 2025, CAFC 2025 Top Frauds, UK Finance Annual Fraud Report 2025, GASA Global State of Scams 2025 Derived metrics: Axis Intelligence Scam Exposure Score™ (SES), cross-agency US combined loss total, extrapolated true-loss estimates, AI-fraud implied-loss range License: Creative Commons Attribution 4.0 International (CC BY 4.0). You are free to share and adapt this data for any purpose with attribution.
Citation format:
APA: Axis Intelligence Research, & Chen, M. (2026, July 17). Online Scam Statistics 2026. Axis Intelligence. https://axis-intelligence.com/online-scam-statistics/
MLA: Axis Intelligence Research and Marcus Chen. “Online Scam Statistics 2026.” Axis Intelligence, 17 July 2026, axis-intelligence.com/online-scam-statistics/.
Chicago: Axis Intelligence Research and Marcus Chen. “Online Scam Statistics 2026.” Axis Intelligence. July 17, 2026. https://axis-intelligence.com/online-scam-statistics/.
Copy this citation block to attribute this research in your work.
Frequently Asked Questions
How much money is lost to online scams globally each year?
According to the GASA Global State of Scams 2025 Report — which surveyed 46,000 adults in 42 countries — approximately $442 billion was lost to scams globally in 2025. This survey-based estimate is the most geographically wide-ranging available; official government complaint figures are considerably lower because only 5–15% of victims ever report to a government agency.
What is the most common online scam in 2025?
By global frequency, shopping scams affected the most people: 54% of all scam victims surveyed by GASA in 2025 encountered a shopping scam. By financial harm, investment fraud — particularly cryptocurrency pig butchering schemes — caused the most damage, generating $8.648 billion in US losses alone, per the FBI IC3.
What is pig butchering and how much does it cost victims?
Pig butchering (shā zhū pán) is a long-duration investment scam in which criminals build fabricated romantic or friendship relationships with victims before directing them to fraudulent cryptocurrency platforms. Over the four-year period from 2020 to early 2024, pig butchering networks moved more than $75 billion in crypto proceeds from victims, per blockchain research by University of Texas professors Griffin and Mei. The average US crypto investment fraud victim in 2025 lost approximately $117,600.
Which country loses the most money to scams?
In absolute reported losses, the United States leads, with $20.877 billion reported to the FBI IC3 in 2025. However, Axis Intelligence Research’s Scam Exposure Score™ — which adjusts for population, reporting infrastructure, and survey-measured encounter rates — shows that when per-capita exposure is considered, Australia and the US are closely matched as the most scam-exposed nations among those with mature reporting systems.
Are online scams getting worse?
Yes, across every jurisdiction with longitudinal data. The FBI IC3 has recorded five consecutive years of loss growth; 2025 marked the first year in the agency’s 25-year history that complaint volume exceeded one million. GASA’s global survey shows 57% of adults worldwide encountered a scam in 2025 — up from earlier years. AI is accelerating scam sophistication, with the UNODC documenting a 600% surge in AI-generated deepfake fraud content in Southeast Asia in 2024 alone.
Who is most at risk of losing money to online scams?
Older adults suffer the highest per-victim financial losses: Americans aged 60 and older averaged $38,500 per reported loss in 2025, compared to the overall IC3 average of $20,699. However, younger adults (18–29) have the highest social media scam exposure rate, and middle-aged adults (40–59) collectively account for the largest share of crypto investment fraud losses by age cohort, per FBI IC3 data.
What are imposter scams and why did losses triple since 2020?
Imposter scams involve criminals posing as trusted entities — banks, government agencies, tech companies, or even family members — to convince victims to transfer money. The FTC recorded $3.5 billion in imposter scam losses in 2025, nearly three times the 2020 figure. Losses have grown because AI and social engineering tools have dramatically improved impersonation quality: fake bank security alerts, voice-cloned phone calls, and deepfake video calls are now documented fraud delivery mechanisms.
How do I report an online scam?
In the US: file a report at ic3.gov (FBI) or reportfraud.ftc.gov (FTC). In Australia: report to Scamwatch at scamwatch.gov.au. In Canada: file with the Canadian Anti-Fraud Centre at antifraudcentre-centreantifraude.ca. In the UK: report to Action Fraud at actionfraud.police.uk. Reports — even when no money was lost — contribute to intelligence databases that help identify criminal networks.
What is the Axis Intelligence Scam Exposure Score?
The Scam Exposure Score (SES)™ is a proprietary composite metric developed by Axis Intelligence Research to enable cross-country scam environment comparisons. It combines four factors — scam encounter rate, money-loss rate, per-capita reported losses, and reporting detection gap — each normalized on a 100-point scale. The US scores 88 (Critical), the highest of any tracked nation; Australia, Canada, and the UK all score in the High tier (65–79). The SES is an Axis Intelligence Research original metric not available from any national reporting agency.
Why does the same scam have different names in different countries?
Jurisdictional reporting taxonomies were developed independently and reflect local law enforcement classification systems. What the FBI calls “investment fraud” largely overlaps with what the ACCC calls “investment scams” and what the CAFC calls “investment fraud” — but not perfectly. “Pig butchering” spans multiple FBI categories (investment fraud and romance/confidence fraud) depending on which phase of the scam triggered the complaint. “Authorized push payment fraud” (UK terminology) has no direct FBI equivalent. These taxonomic gaps are why cross-country dollar comparisons require methodological caution.
What is being done internationally to combat online scam networks?
Several coordinated responses launched in 2025–2026. The US DOJ established the Scam Center Strike Force in November 2025, seizing more than $580 million in digital assets by February 2026. The GASA convened its annual summits in Europe (June 2026) and is scheduled in San Francisco (September 2026) and Bangkok (November 2026). Australia’s National Anti-Scam Centre continued its multi-agency coordination model. The UK’s mandatory APP reimbursement scheme — effective October 2024 — was the first regulatory requirement anywhere in the world forcing banks to compensate most scam victims by default.