On November 25th, the globally leading identity verification and anti-fraud company Sumsub released its annual “2025-2026 Identity Fraud Report”, revealing that due to the application of artificial intelligence, digital fraud has entered a new stage. The report shows that fraud cases are increasing significantly driven by more complexity and professionalism.
Based on millions of verification checks globally and over 4 million fraud attempts, the report also referenced insights from Sumsub’s “2025 Fraud Risk Survey”, which gathered feedback from over 300 risk experts and 1,200 end users.
In 2024, the rise of fraud service platforms and ready-made toolkits made identity crime more “popular”, enabling even non-tech-savvy fraudsters to easily engage in fraudulent activities. In 2025, this trend evolved into a “complex transformation” where fraud numbers decreased, but the level of professionalism increased to create larger-scale losses.
1) “Complex” fraud cases increased by 180% annually: multi-step attacks in all identity fraud cases rose from 10% in 2024 to 28% in 2025.
2) In 2025, 40% of surveyed companies and 52% of end users reported experiencing fraud; 75% believe that fraud activities will increasingly rely on artificial intelligence.
3) Between 2024-2025, the overall identity fraud rate reduced from 2.6% to 2.2%, yet remained higher than the 2.0% rate in 2023.
4) A new trend emerged in forged documents: the proportion of AI-assisted forgeries increased from 0% in 2025 to 2%, primarily driven by tools like ChatGPT, Grok, and Gemini.
5) Industries most impacted by fraud in 2025: online media and dating (6.3%), financial services (2.7%), cryptocurrency (2.2%), professional services – consulting, legal, accounting, marketing, and freelance platforms (1.6%), and video games (1.6%).
6) Primary first-party fraud methods (where the verifier is the fraudster): synthetic identities (21%), payment abuse (16%), application fraud (14%), deep fakes (11%), money laundering (11%).
7) Most common third-party fraud methods (external attackers using or impersonating real users): identity theft (28%), account takeover (19%), credit card testing (17%), social engineering (16%), bot attacks (12%).
8) Overview of the Asia-Pacific region: one in four respondents has been recruited for money laundering; about 80% have heard of the term “money mule” but lack understanding of actual legal and financial risks. Cambodia (17%) has the highest proportion of approved applicants related to fraud networks in the Asia-Pacific region.
9) Regional fraud rates: Europe (-14.6%) and North America (-5.5%) decreased, Africa (+9.3%) increased, Asia-Pacific region (+16.4%) increased, Middle East (+19.8%) increased.
Highest fraud rate regions: Iraq 9.7% (Middle East), Pakistan 5.9% (Asia-Pacific), Tanzania 5.0% (Africa), Argentina 3.8% (Latin America), Latvia 3.7% (Europe), and the United States 1.4% (North America).
As shown in the “Evolution of Fraud Complexity” report, while fraud rates in Europe and North America may be decreasing, their sophistication and impact continue to grow.