Opinion Post – KYC, AI and Fraud
2018, saw $1.5 billion lost due to fraudulent incidents in the world. With over 500,000 identity theft-related frauds being reported, combined with a bulk of unreported incidents ID theft is one of the top 3 types of fraud affecting the banking and financial sector in the world.
FATF and Anti-Money laundering laws
Considering how fraud was affecting the financial system and its negative impact on the economy Financial Action Task Force (FATF) was constituted in 1989, an inter-governmental body whose main objective was to set standards and promote effective implementation of legal, regulatory and operational measures for combating money laundering, terrorist financing and other related threats to the integrity of the international financial system. FATF issued its first recommendations in the year 1990 and was revised in the year 1996, 2001, 2003 and recently in the year 2014.
Following these recommendations, governments across the world including India create laws and regulations on how the financial institutions could acquire customers, verify them and process transactions.
India PMLA guidelines under RBI is one such regulation. Within these guidelines were the processes to follow before onboarding a customer and providing services to them i.e. allowing the customers to transact through the banking systems.
Hackers, fraudsters and the Internet
The advent of the internet meant digitization of everything from paying bills to talking to doctors from buying groceries to opening & operating a bank account. This also meant the rise of internet-based frauds, identity or otherwise it meant people could gain unauthorized access to banking systems, and misuse the information to cause harm to the company, people and sometimes the entire economy of the country itself.
Cyber fraud, KYC and Security Analysts
With cyber fraud and money laundering on the rise around the globe, governments and organizations like FATF including private companies started taking steps towards securing their data both online and offline.
(this is where data security analysts and consultants came into play)
KYC or Know your customers was one way to identifying the customers or users in the banking systems, this allowed governments and organizations to keep track of fraudulent transactions from within their systems and helped them pin-point the sources of money laundering and avoid identity-related fraud to an extent.
Though KYC was helpful in minimizing the cases of fraud and sometimes detect them in advance, the process itself is operations-driven leading to human errors and frauds taking place in cohesion with the employees.
State of KYC now
As internet set in there was a definite increase in identity theft and related financial frauds but it also meant the technology that was used to curb such incidents were developed, including jobs like data security officers and programmer who built out algorithms and encryption that made sure the data within the system remained secure.
And with digitized identity cards, secure barcodes and QR codes it became easier to authenticate the ID information shared with the organization. This meant the advent of digitized KYC software dedicated to making KYC easier and more secure than it was ever before.
In recent times, the Know Your Customer (KYC) has also aligned with Artificial intelligence and machine learning, to automate the verification process itself, technologies like liveness checks and biometric face match is ensuring that person submitted the documents and the person on it are one and the same.
How does it work?
KYC is basically the collection and collation of customer data, i.e. the customer provides proof of identification, proof of address, and a photograph of the customer, to make sure that the person is who he says he is.
Usually, a re-KYC is required, to ensure an updated database of the customer in areas where they might have been a change -for instance, address or marital status or incase there was a minor mistake in the data.
Though in most cases this remains a manual process where an agent visits the customer or the customer visits the company to verify their details. Mainly because someone needs to eyeball the details and look at the customer before they approve KYC.
However, newer and faster ways of getting KYC done are being implemented with the advent of AI and ML gradually taking over the legacy systems. So instead of having an agent visit the customer to manually check the details, more efficient ways like;
- Aadhaar Offline KYC (Processing KYC without the use of biometrics)
- Electronic KYC (Accessible to only customers with Aadhaar number)
- Central KYC
- Video KYC. This involves capturing all details and identification via a video.
Are being used extensively by the BFSI sector to automate the process and avoid any human errors that usually arise in human dependent systems.
The Video KYC works through online submission of a prerecorded video, or by connecting via live videos to send documents required for the KYC process to the banks.
In the case of prerecorded video, documents and identity files sent will be validated with the use of Face match technology, then the video can be scrutinized to ensure that they are authentic and have not to be tampered with.
Minimalist, Oddball Artist , Experimental Chef, Techie By education, Traveller who hates driving, and Named after a bathing Soap(yes, there is a soap called Vigil) and learning to Code. Obsessed withinternational expansion and reading 100 books a year.