Commonwealth released key insights gleaned from a three-year project supported by JPMorgan Chase designed to achieve the promise of artificial intelligence for all by focusing on ensuring that the needs of households earning low to moderate incomes (LMI) are understood and integrated into financial design and implementations.
At this critical inflection point in scaling AI, it is urgent to research and identify the ways in which AI can positively impact the financial lives of these populations, which are disproportionately Black, Latinx and women-led. The report also sheds light on the most promising future directions for technology development.
Historically, the design of new technologies has not fully considered the potential benefits and uses for households earning LMI. As a result, innovations in AI may deepen inequity by not considering the impact they can have for consumers of all backgrounds. Much of Commonwealth's Emerging Tech for All (ETA) research focus has been on conversational AI, and providing actionable guidance to financial service providers on how this technology can support the financial security of U.S. workers earning LMI.
The new report offers key insights, including:
Conversational AI provides a crucial opportunity to improve access for households living on LMI, who are nearly twice as likely to want to bank in person yet have significantly lower rates of access to local bank branches. Conversational AI can provide the personalized and context-sensitive support this group seeks at scale in a way that has never been possible before.
There is a significant untapped opportunity for providers to leverage chatbots to improve financial outcomes. Commonwealth research found that 57% of users in a field test study said using chatbots improved their financial situation. The research also revealed that people had a strong interest in credit building, budgeting, and debt management.
Field test respondents said they valued the judgment-free space with a chatbot for asking sensitive financial questions without shame. Participants also preferred to use a chatbot to resolve straightforward, and commonly experienced actions with easily defined outcomes. Conversely, participants preferred speaking with a human representative when they knew their issues were more complex, potentially involved higher stakes or risk to their finances, or had fewer certain outcomes. However, there is a significant demand for more capable chatbots that can assist with banking actions beyond providing information. These expanded capabilities will be key to providing value for and building trust with customers living on LMI.
Financial institutions have an opportunity to build trust with customers earning LMI through transparent data policies, reassuring branding and messaging, and maintaining the connection to a human agent as a backup option.
"Our world is being transformed by artificial intelligence. For those of us working to build financial security for people living on low and moderate incomes, it's vital - and exciting and challenging - to engage deeply with AI technology," says Tim Flacke, Executive Director at Commonwealth, "We need to research and study it, test and experiment, and find the ways it can foster a more inclusive, effective financial system for everyone."
The report also highlights key opportunities for generative AI to support households living on LMI through the workplace (e.g., workplace benefits), public benefits programs (e.g., eligibility determination), and AI copilots and assistants. There are promising opportunities for generative AI to support back-office processes and customer service agents throughout the financial services industry, saving time and increasing the efficiency of scaling.
"Technology can be a powerful tool to help people build financial health and resilience, especially when it proactively considers the needs of users across the economic spectrum, including those living on LMI," says Miriam Freeman, Vice President, Global Philanthropy, JPMorgan Chase. "The insights surfaced provide a strong foundation to build equitable tools that support a more inclusive financial system."
Commonwealth's ETA research includes three national surveys, interviews with 16 providers, and two field tests with several hundred recent chatbot users, in partnership with fintechs Posh.ai and Boost.ai, and credit unions MSUFCU and ACU of Texas. Research completed in the final phase of the initiative was also used to update Commonwealth's Financial AI for Good Guide & Chatbot, which provides actionable guidance for financial service providers using emerging technologies like conversational AI (chatbots) and automated financial management software to serve populations living on LMI. The initiative also published three podcast episodes with financial technology leaders.