The financial crisis introduced new banking restrictions such as the 2010 Dodd-Frank Act and the Financial Stability Oversight Council. Since then, financial institutions have been trying to find new ways to squeeze cost and improve profits. One way they have achieved both goals is through artificial intelligence. In this article, we'll look at three main areas where banks and credit unions are using AI.
Chatbots - Chatbots are customer facing AI assistants that help customers with a number of tasks. These include finding specific transactions, fetching a balance, answering questions about bill due dates and more. They aren't meant to replace human assistants but do free up employee time for more involved tasks.
Some examples of chatbots currently in use include Bank of America's Erica, UBS's "Ask UBS" and Wells Fargo's chatbot that also integrates with Facebook Messenger.
Automated Fraud Detection - In a February 2018 report produced by the Center for Strategic and International Studies (CSIS) and McAfee, they found that cybercrime cost the world 0.8 percent of global GDP. To put that percentage into perspective, that's equivalent to $600 billion or more than the GDP of Sweden in 2017.
Being able to detect potentially fraudulent transactions and notify the customer so they can take action is critical in avoiding bigger problems. Through pattern recognition, AI can flag a suspicious looking transaction and send an email, text, or even automated phone call to the customer. The customer can then decide to clear the transaction or report it as fraudulent, which bank employees can take further action on. No bank employees need to be involved until the customer reports the transaction as fraudulent.
Automation of Internal Processes - JPMorgan's COIN is a chatbot, but unlike other banks, it is internally facing rather than customer facing. COIN focuses on automated back-office processes such as analyzing complex contracts, handling employee password reset requests and parsing messages from employees.
In a 2017 McKinsey report, they see AI automating 10 to 25 percent of work across banking functions.
AI is all about improving efficiency. AI will supplement rather than replace most jobs. The end result will be new services that haven't yet been introduced or even realized.