In 2018, the financial services industry is using artificial intelligence (AI) and machine learning to drive greater speed and accuracy across business processes. Using AI and machine learning technology, financial services companies are reducing risk, managing fraud, optimizing investment strategies, improving operational efficiency, and delivering more personalized customer service, at scale.
Why is the financial industry seeing so much benefit from machine learning and AI? Consider the process-rich business environment and repeatable, transaction-based interactions that banks have with customers and partners. These interactions create huge, rich data sets around customer behavior and preferences. With AI and machine learning as a crucial piece of their technology strategy, leading companies can interpret and act on data in real-time, to both mitigate risk and provide the most relevant customer experience.
Here are the four ways top financial services companies are using machine learning and AI to drive business growth.
Manage risk and fraud with more intelligent machines
Compared to other industries, analyst firm IDC predicts that banking and retail will invest the most in AI between 2017 and 2020. In banking, much of that capital will be spent on managing risk and fraud. IDC claims that most of the $3.3 billion spent by the industry on AI in 2018 alone will be spent on automated threat intelligence and prevention systems, fraud analysis and investigation, and program advisors and recommendation systems.
Machine learning and AI can help banks and other financial institutions detect irregular customer behavior and flag suspicious transactions. While simple automated rules can often create false positives for fraud or abuse, more complex machine learning models can better predict irregularities based on a customer’s typical patterns of behavior and known edge cases.
In one bank’s case, its AI detected fraud from an identify thief falsely logged in as a customer. The criminal was detected when he used the scroll bar because the real user preferred to use a trackpad when banking. The AI was able to flag this inconsistency because it could interpret massive amounts of data simultaneously.
Make AI-enhanced investment decisions
Machine learning can also help banking and FinServ companies save on operational costs and increase operational efficiency by using AI in investment stress testing. Risk managers surveyed by McKinsey cite 25-50% faster credit decision times from using AI.
Assessing risk is many banks’ top priority when investing in machine learning technology. According to a report from the Financial Stability Board (FSB), one global investment bank uses a machine-learning algorithm to interrogate unlabelled data. This algorithm finds patterns to determine how much capital it has in play at any moment, and how much it can afford to lose.
Bank of America has also seen the potential of using AI algorithms to analyze massive amounts of trade data. All firms have access to the same gigantic amounts of daily data around other companies’ investments — where they’re winning and losing. Bank of America believes the next big opportunity will be in capturing these events using AI, and using the massive data set to better predict investments. This area is still untapped, but the firm expects to soon be able to harness real-time analytics and insights on these information streams.
Automate processes to get more efficient
McKinsey & Company research has outlined how the next wave of AI and automation will transform financial services over the next few years — and companies that use machine learning to improve business processes will reap clear competitive benefits.
The firm predicts that machines will take over 10-25% of human work across a range of functions in the FinServ industry. For example, JP Morgan is using bots to handle 1.7 million internal IT requests this year — the equivalent work of 40 full-time staff members.
Meanwhile, the Australia and New Zealand Bank (ANZ) is using robotic process automation (RPA) so its employees can do less routine work, and focus on more high-value tasks. It’s applying AI to over 40 business processes, producing more than 30% annual cost savings for certain banking functions.
Improving customer experience, at scale
Customer service bots are the most publicly visible form of machine learning. By automating repeatable tasks at scale, a bot is able to interact with human customers across a range of service needs. The highest profile U.S. example is “Erica”, Bank of America’s customer-facing virtual assistant. Erica uses natural language processing to understand speech, text, and intent — and machine learning to gain insights from customer data that can be turned into advice and recommendations.
Financial services companies who have adopted machine learning to lower investment risk, detect fraud, automate business processes, and drive more effective customer engagement are seeing significant returns on their investments.
For over 20 years, Appen’s data services have helped clients enhance their machine learning solutions and products. To find out more about Appen’s financial services solutions, click here.