In a bid to make it easier for the banking industry to keep track of customer identity, IBM Corp. announced Thursday that Bangalore, India-based financial firm Signzy Technologies Pvt. Ltd. will use IBM Watson’s cognitive computing capability with IBM Blockchain services to develop a digital trust platform.
The two technologies together would be used to assist banks to better follow Know Your Customer, or KYC, the industry anti-money laundering regulations mandated by the Reserve Bank of India. According to Signzy, the platform is expected to speed the identity verification process by as much as 80 percent—potentially taking the total process time from two weeks to two days.
“Cognitive computing has the potential to dramatically simplify digital trust processes throughout the financial services industry,” said Ankit Ratan, co-founder and chief executive officer of Signzy Technologies.
Signzy will use IBM Watson’s cognitive computing technology to take what is normally a manual process involving numerous employees poring over customer paperwork in order to compile them into a computer searchable trail of records. The objective of this process is to provide a customer identity history to help reduce fraud, money laundering, financing of terrorism and other illegal activities.
The platform will also use IBM Blockchain, a decentralized trusted ledger service, to establish a common protocol for secure records to store background checks and identity that will be used to reduce tampering.
To cover as much territory as possible, Signzy’s system will collect both structured and unstructured data about customer identity, including financial history, criminal records, court cases, aliases and government filings, in various formats. The system then automates filtering of records for what is relevant and what is not, for example ignoring divorce proceedings.
Banks will be able to upload both documents and images into the platform, which can interpret text and context using the IBM Watson Document Conversation Service. It will be capable of comparing photos to verify that a person’s documentation matches his or her appearance using the AlchemyVision Service. The system can also ingest video and audio content by applying speech to text.
Documents and analysis stored in the blockchain would become part of a secure trusted ledger designed to both cryptographically protect private customer information and reduce the risk of identity theft and fraud.
As the system is designed to detect fraudulent and illegal activity, Signzy has primed the platform with common examples of both normal behavior and red flags. The expectation is that the system will apply this basic framework to incoming data in order to identify anomalies and generate a “forgery score.”
Furthermore, IBM Watson’s cognitive computing capability will apply machine learning processes based on this historical priming to current incoming data giving it the ability to detect as-yet unknown types of fraud, financial trickery or criminal behavior.
The company also hopes that because it is using machine learning to provide the verification process, it should mitigate the risk of human error or corruption.