The financial sector and other industries have long relied on the “know your customer” (KYC) authentication mechanism to confirm a person’s identity when opening an account and periodically confirming that identity over time. However, with the emergence of generative artificial intelligence (AI) using large language models (LLMs) to create highly persuasive document replicas, many security executives are reconsidering how KYC should operate in a generative AI world.
Generative AI uses LLMs to enable KYC fraud in various ways. For example, someone could walk into a bank with AI-generated replicas of documents, such as a driver’s license and a passport, making it difficult for bank staff to authenticate the documents. Furthermore, generative AI can create bogus documents quickly and on a massive scale, posing a significant security threat. Cyber thieves could take advantage of this technology to perpetrate unlimited fake account setup and account recovery attempts.
The threat of generative AI doesn’t just stop at creating false documents. Lee Mallon, the chief technology officer at AI vendor Humanity.run, is concerned that thieves could use LLMs to create fake personal histories that validate AI-generated fake KYC documents. This could involve creating elaborate backstories for fraudulent identities, making it challenging for banks and government agencies to verify the authenticity of individuals.
Moreover, the growth of fake identities using generative AI poses a significant challenge for KYC techniques. Alexandre Cagnoni, director of authentication at WatchGuard Technologies, believes that KYC processes will need to incorporate more sophisticated identity verification processes, including the use of AI-based validations and deepfake detection systems. The lack of advanced deepfake detection technologies presents a challenge for financial institutions, as they will need to invest in robust systems to combat the growing threat of fake identities.
As such, the use of generative AI and LLMs raises concerns about the security of traditional KYC processes. Implementing AI-based validations and deepfake detection systems will be essential for financial institutions and other industries to mitigate the risks posed by the advancement of generative AI. The implications of this technology extend beyond document replication, highlighting the need for enhanced identity verification processes to combat the growing threat of KYC fraud.