Digital forgeries account for 47% of document fraud detected in Asia Pacific (APAC), while regional deepfake attacks increased 116% year on year, according to Entrust’s 2027 Identity Fraud Report. Those trends point to growing risks for digital identity verification as criminals manipulate document images and use AI-generated content to impersonate legitimate users.

Table of contents

Covering identity fraud activity between September 2025 and September 2026, the report draws on data from Entrust’s verification services, which process millions of identity checks annually across 195 countries. Those figures reflect activity within identity verification systems and may differ from broader market trends. To account for differences in its client and industry distribution, Entrust normalised the data.

Digital document manipulation is prevalent in APAC

Among the three regions examined, Entrust identifies APAC as the most digitally native fraud market, with the highest proportion of fully digital, mobile-first onboarding. Unlike counterfeiting, where a fraudulent document is created from scratch, digital forgeries typically involve altering an image of an existing identity document before submitting it for verification.

Even so, most document fraud detected in the region remains relatively straightforward. Entrust classified 74% of APAC cases as easy to detect, while medium-sophistication attacks accounted for 26%. Those more convincing attempts require greater skill and better tools to produce, making them harder to identify.

At the global level, however, medium-sophistication identity fraud rose nearly 20-fold and advanced identity fraud increased 268%. The figures indicate a rise in attacks requiring more sophisticated techniques, even as simpler forms of document fraud remain common in APAC.

Earlier data from Entrust’s 2026 Identity Fraud Report, covering September 2024 to September 2025, showed that national identity cards accounted for 60% of fraudulent document submissions in APAC. Globally, Bangladesh national ID cards were the most targeted document type in 2025, with older paper-based identity documents identified as particularly attractive to fraudsters because they can be easier to reproduce.

Deepfakes and injection attacks target verification systems

Global deepfake attacks rose 165% in 2026, with Entrust detecting an average of one attempt every three minutes. Alongside its 116% year-on-year rise, APAC recorded the second-highest share of deepfake activity among the three regions studied.

In the preceding 2026 report, deepfakes featured in approximately one in five biometric fraud attempts. Criminals commonly relied on generating synthetic faces, replacing a person’s face in recorded or live video, and animating static photographs to imitate a real user.

Deepfake attacks are especially prevalent in high-risk financial services:

Deepfake attacks in high-risk financial services The infographic shows Crypto at 60%, Digital-First Banks at 22%, Payments & Merchants at 13%, Investing at 2%, Remittance & FX at 2%, and Travel at 1%.
Deepfake attacks in high-risk financial services Crypto 60%, Digital-First Banks 22%, Payments & Merchants 13%, Investing 2%, Remittance & FX 2%, Travel 1%. The category values are listed beneath the chart.
  • Crypto 60%
  • Digital-First Banks 22%
  • Payments & Merchants 13%
  • Investing 2%
  • Remittance & FX 2%
  • Travel 1%

Outpacing deepfake growth, global injection attack volume climbed 224% in 2026. Rather than passing through a device’s physical camera, these attacks feed fraudulent images or videos directly into identity verification software. Using virtual cameras and device emulators, criminals can make stolen or AI-generated identity media appear as though it was captured live, allowing manipulated content to enter verification systems without normal camera capture and potentially undermining checks designed to establish whether a genuine person is present.

On the defensive side, Entrust’s earlier report identified several detection methods, including examining device and video input signals, introducing randomised actions during live biometric checks, and analysing submitted content for inconsistencies. It also referenced CEN/TS 18099 as an emerging standard intended to help evaluate injection-attack detection capabilities.

Number of Injection Attacks
Number of Injection Attacks Line chart showing Suspected Rate (%) from September to August. The purple line represents 2023-2024 and the magenta line represents 2024-2025. For 2023-2024, the monthly values are 0.14%, 0.12%, 0.13%, 0.10%, 0.15%, 0.15%, 0.16%, 0.26%, 0.30%, 0.32%, 0.26% and 0.28%. For 2024-2025, the monthly values are 0.32%, 0.37%, 0.26%, 0.27%, 0.29%, 0.30%, 0.32%, 0.28%, 0.26%, 0.27%, 0.30% and 0.29%.
Suspected Rate (%)
Month 2023-2024 2024-2025
September 0.14% 0.32%
October 0.12% 0.37%
November 0.13% 0.26%
December 0.10% 0.27%
January 0.15% 0.29%
February 0.15% 0.30%
March 0.16% 0.32%
April 0.26% 0.28%
May 0.30% 0.26%
June 0.32% 0.27%
July 0.26% 0.30%
August 0.28% 0.29%
0.05% 0.40%
2023-2024 2024-2025

Beyond individual attack techniques, the 2026 report examined the organised operations supporting identity fraud. Entrust identified 53 distinct fraud rings since 2023, including four that targeted multiple clients across different sectors. Fraud-as-a-service platforms further give criminals access to techniques that can support large-scale attacks by selling ready-made scripts, device emulation tools and credential lists.

Simon Horswell, Senior Fraud Specialist Manager at Entrust, warned that successfully bypassing an identity check could expose financial institutions to further fraudulent activity. “That raises the stakes for financial institutions where a successfully passed identity check can open the door to far more than one fraudulent transaction. Banks need to deploy layered defences that can assess the authenticity of identity media alongside the person, device, and behaviour across each interaction.”

Social engineering targets people with account access

Reported social engineering attempts rose 520% in 2026 and increased 75-fold over two years, making social engineering the fastest-growing attack method identified in Entrust’s latest report. Attackers target customers and employees authorised to access accounts, approve transactions or complete security checks. By persuading or coercing those individuals to use their genuine identity credentials, criminals can take control of existing accounts without necessarily defeating the underlying identity verification technology.

Fraud patterns also differ across financial services and other industries, according to Entrust’s preceding report. In its September 2024 to September 2025 dataset, 82% of payment-sector fraud occurred after onboarding, compared with 64% in professional services and 55% in digital-first banking, whereas cryptocurrency platforms faced a different pattern, with 67% of fraud attempts occurring during account opening.

Entrust recommends extending verification and fraud detection across account opening, account recovery, authentication and high-risk transactions. Recommended measures include biometric verification, device intelligence, injection-attack detection and cryptographically verifiable credentials.

The report also identifies developments in digital identity programmes across APAC, including Malaysia’s electronic Know Your Customer (eKYC) requirements, the Philippines’ PhilSys national identity system and Australia’s Digital ID framework. Across the region, governments are expanding these programmes and updating verification requirements to improve trust in digital transactions.

Share