Financial Services is a Natural Match for AI

Artificial intelligence has been the stuff of science fiction for decades – “I, Robot” was a collection short stories in the 1950s and a movie more than 50 years later – as well as a pursuit for just as long by technology and scientific communities.

With the rapid rise in computing power in recent years combined with more sophisticated algorithms and massive amounts of data being generated, AI and its subsets – including machine learning, deep learning and natural language processing – are gaining traction in both the consumer and commercial worlds. They create the ability to crunch through huge amounts of big data to find patterns and gain insights in real-time to drive more timely and better business decisions in a way that humans can’t. The technologies also can learn as they go, improving the information they derive from the data.

AI in business is still in the early stages of adoption. A Gartner survey of CIOs found that only 4 percent of chief information officers have implemented AI, though another 46 percent have created plans to do so. However, it’s a different scene in financial services, which has been at the forefront of adoption. In a report by Adobe and Econsultancy, 61 percent of industry executives are using or will adopt AI within 12 months. The most popular uses are in data analytics, optimization, testing, automated campaigns and on-site personalization.

Financial services firms gather huge amounts of data on customers and their businesses and can leverage insights from that data to gain a competitive advantage. They also can use AI to create services – think of chatbots – that can vastly improve the user experience.

Where AI is being used in financial services

• Fraud detection and cybersecurity. With AI, banks can become more proactive, running through large amounts of data to detect anomalies that can indicate fraud or other potentially harmful activities, and then can alert administrators and security personnel. The technology also can quickly rule out false positives.
• Investments and portfolio management. A common term in financial services today is “robo-adviser.” Robo-advisers collect data from users – from their age and income level to financial goals and the amount of risk they’re comfortable with – and then analyze the date using AI algorithms that then develop financial portfolios and investment plans the user can leverage.
• Compliance. Regulations governing everything from money laundering to privacy to investing are complex, differ from region to region and are ever-changing. AI- and machine learning-based solutions can learn and remember the myriad regulations and laws and their inevitable changes and ensure that financial services institutions remain in compliance.
• Trading and forecasting. Machine learning and deep learning can analyze financial data and find patterns in stock price movements, historical data and other available information that give investors and analysts greater insights into the behavior of the markets and then can make the best trades based on the information.
• Underwriting loans and insurance. Machine learning algorithms can churn through massive amounts of customer data, from age and marital status to previous loan or insurance results, and the information from the analyses can then be used to detect trends in the market and be applied to future underwriting decisions.
• Customer service. The most notable use of AI for customer service has been the rise of chatbots, which use a combination of machine learning algorithms and natural language processing interfaces that can easily deal with most customer concerns, from transactions and onboarding to more complex issues. They also can leverage other ways of communicating with customers, including texts and social media. Many major banking institutions are already using chatbots, including Bank of America, JP Morgan, Capital One and Wells Fargo.

AI and machine learning also play a crucial role in processing documents, from initial orders and invoices to payments and managing exceptions. AI-based solutions can improve speed, accuracy and efficiencies in processes that still tend to be done manually, and machine learning enables the system to learn from users. Patterns in data can be found more quickly, and process flows are improved. This is important for financial executives, a third of whom still use manual methods like Excel spreadsheets for such tasks as data collection, calculations and reporting, according to the Institute of Finance and Management.

More automated document processing can enable CFOs to more easily view organizational spending, accounts payable cash flow and other aspects of a company’s finances. 

Adopting AI in your environment

Many financial services organizations have already picked up the AI ball and are running with it. However, to get the full benefits of the technology, there are some things to keep in mind:

• Make sure the data is clean. AI algorithms are only as good as the data that goes into them. The cleaner the data, the better results you’ll get, and clean can mean a lot of things. For example, data is kept in different tables and databases throughout a company, and a customer address listed in one system may use “Street” while another one uses “St.” Customer names in one place may use a middle initial, but not in another. The data needs to be consistent and tagged correctly.
• Don’t swing for the fences immediately. Start with smaller projects that target specific pain points and have attainable goals, such as saving money through automating accounts payable and receivables, creating efficiencies, reducing costs or improving customer service. Such projects also will allow provide lessons that can be applied to larger, more complex implementations.
• Use due diligence in choosing an AI vendor. “AI” and related terms are increasingly being attached to every hardware and software product and service. Make sure the vendor you choose can tell you how their AI solutions reach the findings they do, which is particularly important in such highly regulated industries like financial services. Also, have the vendor explain how their offerings will address your challenges, the services and support they bring to back them up and the ROI you’ll get from choosing them.
• Don’t forget the people. AI solutions will have an impact on employees, from how they work to whether they’ll still have jobs. Companies need to ensure that employees have the skills needed to implement and manage AI solutions in the office. In addition, organizations need to engage workers at the beginning of the AI journey to gain their support and buy into the understanding that the technology can enhance their work.
• Be grounded in your expectations. AI- and machine learning-based systems aren’t magic bullets. They can collect, store, analyze and act on massive amounts of data, giving insights, recommendations and visibility into your environment, but humans need to make the decisions based on all that.

The data-rich financial services industry has become a leader in the adoption and use of AI technology and it’s being used in a variety of ways. It’s protecting against fraud, driving investments and improving customer service, and as the solutions evolve and organizations become more comfortable with it, the use of AI and its various subsets will only grow. Those that embrace it now and learn to harness its power will see a huge advantage over their lagging competitors.

Eric Bussy is product management director at Esker.