HDFC Bank banks on AI to take efficiencies to new level
HDFC Bank has always been a bank of scale. Once the bank is convinced of an emerging technology, it rapidly scales it up across the organization. Take the case of AI. EVA, its AI-powered chatbot, was used by the bank to just answer basic customer queries. Today, the bot can handle transactions on like bill payments, ticket booking, etc. Eva has already answered more than 5 million queries with more than 85% accuracy. The chatbot holds more than 20,000 conversations everyday with customers from all over the world.
“Our intent is to embed AI in every business process. We see huge potential for AI and its impact for radically improving the efficiencies,” states Munish Mittal, Group Head-IT & CIO, HDFC Bank. A case in point is the signature verification process for clearing the cheques. The bank is using machine learning with almost 80% accuracy to process more than 2.5 lakh cheques a day.
The bank also uses an intelligent AI-enabled bot which is used for improving the efficiency of the credit decision process. Eye balling is of paramount importance in the credit decision process, and the bot mimics a human by eye balling the customer demographics for those who have applied for loans against the already existing base. The bot can crosscheck multiple columns for matching and cross checking it with available information. This parallel processing ability of the bot to work on ten probable matches at once, instead of a human doing it one by one, improves efficiencies by a huge level.
Munish Mittal says that AI is now being looked for automating every possible process. “We are looking at using AI for understanding the preferred channel that the customer wants to be reached out or be connected.” The bank is also using AI for processes such as underwriting, credit score and for credit card payment checking. For improving its cyber security posture, the bank is looking at AI for anomaly detection in its Security Operations Center.
If you have an interesting article / experience / case study to share, please get in touch with us at [email protected]