Top 5 AI stories of 2023
Bank of America, Fifth Third, TD Bank, Ally Financial explore AI
AI and generative AI uses exploded in 2023 and the financial industry felt the ripple effects.
Ninety-nine percent of financial services industry leaders have deployed AI and are planning to use the tech across their operations, according to a Dec. 11 report by EY. Of 300 financial services executives surveyed at companies with revenues of more than $2 billion, 77% view gen AI as an overall benefit to the financial services industry.
This year, financial institutions implemented AI and generative AI into their operations to streamline both internal and client-facing chatbot experiences.
Here are Bank Automation News’ top five AI-related stories of the year:
1. Bank of America deploys AI assistant Erica internally
Bank of America deployed its AI-driven chatbot, Erica, internally in May to save bankers “hours of research and manual work,” Jorge Camargo, senior vice president of digital product management, told BAN in May.
As Erica adds efficiencies internally, the chatbot’s consumer adoption is also ticking up, according to the $2.4 billion bank’s Q3 earnings report. During the quarter, Bank of America reported that consumer adoption of Erica increased 27% year over year to 170 million interactions.
2. Fifth Third updates chatbot for fluency
Fifth Third Bank looked to ChatGPT to enhance the capabilities of its chatbot’s natural language understanding model in September.
The bank’s chatbot, Jeanie, facilitates around 200,000 conversations per month and can understand close to 30,000 customer phrases and queries, Michelle Grimm, senior director of conversational AI at Fifth Third, previously told BAN. The chatbot knew about 35 phrases when it was launched in 2020.
Fifth Third increased its tech spend 8% YoY in the Q3 to $115 million and is modernizing its platform to better serve its customers and make operations more efficient, according to its earnings supplement.
3. TD Bank explores AI for cross-selling
TD Bank is exploring the use of AI to predict consumers’ needs and use the technology to cross-sell its products.
AI will help TD in identifying targeted offerings toward a customer that will help it bring the customer deeper into its ecosystem and increase revenue for the bank, Chris Giamo, executive vice president and head of commercial banking, told BAN in September.
The $386 billion bank aims to use existing customer data to predict what issues they might run into or what financial product might help them, Giamo said, noting that if a TD Bank commercial customer accepts payment in checks or cards, the bank can suggest an upgrade to its payments channels or provide merchant services to them.
4. Inside look: Discover Financial Services’ approach to responsible AI
Card giant Discover Financial Services is using static and supervised machine learning algorithms to train its models to remain compliant and develop AI in a responsible manner.
Static machine learning (ML) consists of models that are trained offline, Arjun Kannan, director of data science research, told BAN in June.
For certain operations, Discover uses models that are not trained with real-time data, Kannan said.
“Why that’s important is when it comes to compliance,” he said. “It’s hard if you are constantly updating with real-time information and making decisions with rapidly updating models.”
Ally Financial has developed an AI playbook to help its employees understand the technology and aid in the development and deployment of AI across the organization.
Employees can pitch ideas to Ally’s AI development team about what features they would like to have in their department and existing limitations of the technology, Sathish Muthukrishnan, chief information, data and digital officer at Ally, told BAN. Ally is using Ally.ai to help assist its customer relationship team and uses gen AI for targeted marketing purposes.
Get ready for the Bank Automation Summit U.S. 2024 in Nashville on March 18-19! Discover the latest advancements in AI and automation in banking. Register now.
