Using biometrics to counter the growing threat of fraudulent identities

Synthetic identity fraud, the creation of new identities using stolen or made-up data, is the fastest-growing financial crime in the United States. This type of fraud is on the rise due to the easy availability of personally identifiable information on the dark web and advancements in generative AI that can produce realistic images and videos.

The complexity of synthetic identity fraud is increasing with the growth of digital interactions and technological advancements. Fraudsters often create detailed personal information by mixing real and fabricated information, often using social security numbers from children or the recently deceased.

These synthetic identities, nurtured over months or years, may have credit scores just below the threshold considered "good." The average payoff for such fraud is estimated to be between US$81,000 and US$98,000, with some attacks resulting in the theft of millions.

Detecting synthetic identity fraud is challenging because fraudsters establish a comprehensive public history using fabricated credentials. Approximately 85% of synthetic identities in the emerging consumer sector evade third-party risk models. 

To address this, banks and financial institutions should adopt rigorous model risk management procedures and leverage external tools that utilize deep learning to analyze various characteristics and data points of users' identities.

Biometrics, both physical and behavioral, offer a broader safety way against synthetic identity fraud. Physical biometrics analyze unique traits such as palm vein patterns, retina details, vocal pitch, and ear canal shapes. However, these systems can be susceptible to creative workarounds, such as the recent hacking of facial identification technology with glasses. Behavioral biometrics, on the other hand, provide continuous authentication by tracking dynamic information about users, such as touchscreen behavior and typing habits.

To counteract fraudulent actions, financial institutions should focus on refining "liveness detection" checks and invest in powerful biometric tools that assess various indicators simultaneously. 

Behavioral biometrics, complemented by nonbiometric factors like location histories and spending habits, can effectively spot synthetic identities. Additionally, third-party tools with synthetic or artificial biometric data can enhance accuracy and inclusivity, addressing issues like biased AI systems trained on limited demographics.

As financial institutions extend biometric security to new technologies such as digital currencies, a zero-trust security model becomes crucial. Biometrics, including fingerprint verification, can play a huge role in ensuring security in digital transactions. These systems may also enhance customer experiences, and facilitate financial inclusion by providing secure access to those without formal identity documents.

Biometrics offer a win-win for banks, playing a critical role in fraud detection while contributing to financial inclusion and improving customer experiences.