AI is coming to a bank near you

RBC is opening a new lab in Edmonton and is teaming up with the Alberta Machine Intelligence Institute, a division of the University of Alberta’s computer science department, to drive research in this emerging field.

Scotiabank is providing $1 million in funding over the next three years to a venture called NextAI (which RBC also funds) to promote access to capital, education and mentorship for young AI developers.

AI is based on the idea of machine learning – a branch of computer science that builds systems that can assimilate large volumes of data, then adapt output in response to growing, changing sets of data. This is a developing field that underpins everything from smartphone digital assistants to self-driving cars. There also are key applications for this technology in the financial services sector, including increasingly sophisticated robo-advisors, securities trading and compliance automation.

According to a new report from New York-based Citigroup Inc.: “Advanced analytics and artificial intelligence can be applied in every part of banking, from real-time customer engagement to more efficient operational processing to better risk and fraud management.”

At the front end, AI increasingly will shape the interaction between companies and their retail clients. At the back end, analysts foresee great potential for AI to help automate existing labour-intensive processes, which would enable firms to slash their costs.

A report from U.K.-based research firm Economist Intelligence Unit Ltd. (EIU) states that a handful of firms already are using AI-enabled algorithms to handle portfolio trading – a trend that is spreading to retail investing. Robo-advisors typically use traditional, preprogrammed algorithms to generate portfolio recommendations. Increasingly, however, these systems will employ greater volumes and varieties of data while also becoming more adaptable.

The EIU report states: “Financial management firms are now beginning to apply AI techniques to robo-advice delivery. When fully integrated, the machine-generated recommendations will be based on a wider scale and more in-depth analysis of market and environmental data (including social media sentiment), as well as each individual’s past investment behaviour and preferences. The algorithms will, moreover, evolve and adapt themselves as situations and data change.”

Amid this increasing use of AI in front-line applications such as retail banking and investing, a report from Chicago-based consulting firm Accenture PLC predicts that within in five years, more than half of customers will be selecting services to use based on the quality of a company’s AI rather than on the firm’s traditional brand.

At the same time, AI will help financial services firms slash their costs in some labour-intensive areas. For example, the Citigroup report states that the growing use of AI is likely to enable banks in developed markets to reduce the number of physical branches by 30%-50% from 2014 levels. The number of required full-time employees could drop by similar amounts by 2025, the report adds, mostly as a result of increased automation in retail banking.

The Citigroup report states this trend already is playing out in northern Europe, where banks in Scandinavia and the Netherlands have halved the number of branches from peak levels. U.S. banks are expected to follow suit in the years ahead. According to the report: “The future of branches in banking is about focusing on advisory and consultation rather than transactions.”

The Citigroup report states that two key factors will drive the way in which banks re-engineer their business models in various markets. The first is labour laws. The second is acceptance by bank users in the use of digital services, which allow banks to reduce the number of branches they require without harming market share or customer satisfaction. On that basis, the report ranks Canada among the leaders, along with the U.S., Singapore, Hong Kong and several northern European countries, in terms of both labour market flexibility and digital readiness.

Another big target for cost-cutting is compliance spending: AI has the potential to help slash that expense through the development of so-called “regtech” (regulatory technology). The Citigroup report flags regtech as an “area to watch” in 2017, with its significant potential to cut compliance costs: “Regtech implements risk-management and monitoring systems, which could help financial institutions to identify, measure, monitor and control risk.”

The Accenture report states that AI-enabled technology is being used to address time- and labour-intensive processes such as resolving failed trades. The report cites an example of a New York-based investment bank that adopted a new system that enabled the company to resolve 80% of its failed trades without human involvement, thereby reducing the average resolution time from 47 minutes to just four minutes.

On the regulatory front – for now, at least – Canada’s regulators appear ready to embrace the technology. A comment from the Ontario Securities Commission (OSC) indicates it doesn’t have an official position on the use of AI, but reports that various firms are using AI in areas from investment decision-making to client “onboarding” processes (such as “know your client” requirements). Furthermore, the OSC has no objection, as long as firms meet their overall regulatory obligations.

© 2017 Investment Executive. All rights reserved.