How Artificial Intelligence will impact banking and financial services

H-FARM Consultancy team

The banking sector has always been the driving force behind the application of new technologies: automation has been used for back-office and operations since the 1960s. Investments in Artificial Intelligence have grown during the ‘80s in the form of expert systems. Despite this particular focus on IT, many banks still lack the necessary agility to digitally transform their business, which could include modern AI technologies. According to a PWC study,

2 banks out of 3 in the US have not yet adopted AI technologies due to operations, regulations and budget or resource constraints.

What is AI?

Let’s take a step back and let’s define what Artificial Intelligence really is. If we wanted to apply the 1955 definition by John McCarthy, winner of the Turing Prize in 1971 for his contributions on Artificial Intelligence, the mission of the AI is:

“Making a machine behave in ways that would be called intelligent if a human were so behaving”

AI ​​is very often connected to the concept of Machine Learning: the amount and fragmentation of available data make it necessary for machines to learn and perform increasingly complex tasks in an automated manner.

Machine Learning (ML) and Deep Learning are often mistakenly used as synonyms, although Deep Learning is a type of ML that learns through multi-layer neural networks. People often talk about Cognitive Computing, even though it does not have a very clear definition. You could see how a part of AI ​​focuses on simulating the process related to human thinking based on how the brain works.

Among the many fields related to Artificial Intelligence, our CELI team (now part of the H-FARM Industry division) focuses on the creation of systems with linguistic skills, which are able to automatically understand and use natural language.

How to apply AI and speech technology to banking and financial services?

1 Onboarding phase
The potential customer is not keen to search information through keyword-based search engines or wait a long time over the phone. Natural-language search engines and chatbots (even voice bots) can overcome these issues and allow the customer to search for information like “How to open a bank account” in a simple, immediate, and conversational way.

2 Caring
The goal is to proactively communicate the information customers want before they call the contact center. Again, customers can quickly find the answers they are looking for through a search engine or chatbot (or voice bot) on the website, or on the IVR channel by leveraging the power of voice technology and natural language understanding. In general, this easy-to-use, intuitive, and multichannel tools reduce time and provide customers with the customer experience they want, maximizing it at the same time.

Which banks have implemented AI technologies?

1 Bank Of America
launched ERICA, the first AI example to help customers manage their savings. Erica uses AI, predictive analytics, and conversational interfaces to assist and help customers throughout the customer lifecycle, creating a tailored experience for each customer.

2 JPMorgan Chase
has recently introduced a Contract Intelligence (COiN) platform to analyze legal documents and extract relevant conditions and data. The manual revision of 12,000 annual sales contracts normally required 360,000 hours. The result of initial implementation through machine learning technologies has shown that the same number of contracts can be revised in minutes.

3 N26
is the first European direct bank built following Design Thinking. For some time, the app has been integrated with Pulse26, a virtual contextual assistant for financial analysis which can provide personal insights based on the actual individual consumer needs.

4 CapitalOne
was the first bank to offer a new model of engagement through a completely different channel. By integrating the bank’s APIs with Amazon Echo, customers can ask Alexa (the Virtual Assistant in the device) for real-time information about their bank account and perform operations with the simple use of voice.

5 City Bank
has recently acquired Feedzai, a data science company that works in real-time to identify and eliminate frauds even on online banking. Through a steady and rapid evaluation of a large amount of data, Feedzai can conduct large-scale analysis and identify fraudulent or suspect activity by alerting the customer immediately.

Trends

It seems that big banks, at least in the US, have strong AI applications in development and some of them are already in use. This may also depend on the great difficulty in absorbing and retaining AI and machine learning talents, much more easily “conquered” by Google, Facebook, and Amazon. Surely chatbots and conversational interfaces emerge as a trend inspiring enthusiasm and curiosity in the banking world, even in Europe. According to a survey conducted by Accenture on 600 bankers

76% of bankers think that over the next three years most banks will develop and make Artificial Intelligence interfaces available as their first point of contact for customer interaction.

This scenario certainly poses many questions and doubts about the future of branch employees in direct contact with the end customer.