Veridas introduces new injection attack detection feature for fraud prevention

Imagine you’re watching a film and halfway through, unbeknownst to you, an invisible actor hijacks the plot and begins telling a story that will benefit them. 

This is roughly how injection attacks work: they hack into live audio or video feeds and use Generative AI to take control with synthetic identities, enabling identity fraud in real time. Since they involve both a biometrics aspect and a cybersecurity aspect, they present a unique challenge for fraud prevention. 

The EU published a standard for biometric injection attacks, CEN/TS 18099, last year.  For now, they remain ungoverned by the International Standards Organization (ISO) – although an ISO working group is developing a standard for biometric injection attack detection.

To stay ahead of the curve, Spanish biometrics company Veridas has introduced an advanced injection attack detection capability into its system, to combat the growing threat of synthetic identities and deepfakes. 

According to a recent identity fraud report from the firm, which provides document verification, selfie authentication and liveness detection, 85 percent of financial fraud cases now involve synthetic identities. 

Veridas says that standard fraud detection only focuses on what it sees or hears – for example, face or voice biometrics. So-called Presentation Attack Detection (PAD) looks for fake images, videos and voices. Deepfake detection searches for the telltale artifacts that give away the work of generative AI. 

Neither are monitoring where the feed comes from or whether the device is compromised. 

Injection attacks pump synthetic data directly into identity systems, bypassing device sensors. Veridas’ injection attack detection system addresses this by verifying device authenticity during onboarding, employing liveness detection across face, voice, and document authentication to secure both the biometric input and the device it comes from. 

Javier San Agustín, CTO at Veridas, says leading financial institutions in Italy, Spain, the U.S., and Mexico are already using the tool, which he calls “a game-changer for businesses facing the challenges of Gen AI fraud.”

“By adopting Veridas’ injection fraud detection, companies can protect their operations, maintain customer trust, and confidently meet compliance requirements.” 

Article Topics

biometric liveness detection  |  biometrics  |  cybersecurity  |  deepfakes  |  fraud prevention  |  generative AI  |  injection attacks  |  synthetic data  |  synthetic identity fraud  |  Veridas