Fraudsters are increasingly bypassing remote identity verification systems by utilizing virtual camera software to inject manipulated media directly into the authentication process. According to the 2027 Identity Fraud Report published by Entrust, this method allows attackers to circumvent the physical camera hardware of a target device, feeding pre-recorded or AI-generated video and images into verification software as if they were live, legitimate submissions. This technique, known as an injection attack, is often paired with device emulation software to further mask the origin of the submission.
The use of deepfakes in identity verification has surged, with detection rates climbing 165 percent year over year. Criminals are leveraging AI to generate synthetic IDs, photos, and videos to impersonate legitimate users. While the creation of this content is a significant step, the primary challenge for attackers is ensuring the fabricated media successfully passes the security checks integrated into the verification workflow. Once access is granted, these fraudulent identities are frequently repurposed for money laundering, money mule operations, and subsequent financial crimes.
Entrust’s data highlights that while digital manipulation is growing, physical document fraud remains the most prevalent attack category, accounting for 61.1 percent of fraudulent submissions. However, the nature of these physical attacks is evolving; nearly 39 percent of document fraud is now classified as medium-sophistication, involving high-quality alterations and counterfeits that require advanced inspection methods to identify. Despite the rise in digital threats, the overall rate of document fraud saw a 19 percent decline compared to the previous year.
Regional trends reveal varying tactics across the globe. Europe experienced a dramatic 462 percent increase in deepfake detections, with digital forgeries particularly prevalent in Italy. In the Americas, the focus remains heavily on driver’s licenses, specifically those issued in Texas, California, and Florida, while Latin American fraud is dominated by the targeting of national identity cards, most notably in Chile. The Asia-Pacific region reports the highest proportion of digital forgeries, though it currently maintains the lowest rate of social engineering attacks among the regions studied.
Device intelligence has become a critical component in defending against these threats, with fraud rates identified through this method rising 42 percent year over year. By analyzing the camera, device, and session parameters, organizations can better detect when a submission originates from an emulator rather than a physical camera. Despite these technological defenses, attackers are increasingly turning to social engineering to bypass security measures entirely. By manipulating customers or employees into completing identity checks or approving transactions, criminals can force the system to accept genuine documents and biometric data, effectively turning victims into unwitting participants in the fraud.
The professionalization of these criminal operations is a significant factor in the current threat landscape. Organized groups are now dividing labor to specialize in different stages of an attack, from the acquisition of stolen credentials to the execution of complex identity verification bypasses. This coordinated approach allows for the scaling of operations and the targeting of multiple points within a customer’s relationship with a financial service. As AI continues to lower the barrier to entry for creating convincing synthetic content, the industry faces a persistent challenge in securing the identity lifecycle against both automated injection attacks and human-centric social engineering.





