Much has been written about the impact generative AI technology and services like ChatGPT will have on a broad range of industries, including financial services. But for all of the hype around generative AI, when will consumers have a ChatGPT-like service within their financial services website and mobile app?
Many of the largest financial services firms have announced that they are working on internal and/or client-facing generative AI initiatives. For example, JPMorgan Chase is developing a generative AI investment advisor, Morgan Stanley is piloting multiple generative AI projects, Citi has been working on generative AI projects for at least three years, Visa has launched a $100M generative AI venture fund, etc. As of February 2024, however, there have been only a limited number of financial services firms that have actually deployed a live ChatGPT-like generative AI assistant to support their client experience.
As I’ve learned from working with clients in the financial services industry and talking to peers in the industry, 2024 is shaping up to be the year where generative AI in financial services goes from theory to reality. We will see powerful generative AI assistants starting to appear within consumers’ financial services websites and apps. This matters because the financial services sector currently offers only very basic chatbot assistants running on outdated technology. The transition to generative AI assistants will fundamentally change the way the average consumer manages their money and interacts with financial services firms.
This article will first provide a detailed explanation of why the transition to generative AI assistants will have such a significant impact on the financial services industry. We’ll then discuss the “when” question in more detail and a possible timeline for when different financial services industries will start offering client-facing generative AI assistants.
The financial services industry currently offers consumers basic chatbot assistants. The average financial service chatbot can only handle straightforward account servicing questions.
The average financial services chatbot struggles to explain financial concepts, cannot assist with financial planning and budgeting, and does not provide advice or help with investing. The industry’s chatbots are primarily designed to handle relatively straightforward customer support needs, and are not advanced enough to serve as a true assistant or advisor.
In contrast, generative AI assistants can provide reasonably accurate answers to a wide range of financial questions. Generative AI assistants can explain financial concepts/rules, help with budgeting and financial planning, and even provide high-level advice and recommendations when prompted in the right way. Generative AI assistants can also hold a conversation in a somewhat natural way and can refer back to previous comments and discussions. The stilted rules-based chatbots currently offered by the industry cannot hold a conversation or understand references to previous discussions.
One could argue that client-facing generative AI assistants will create the first real “robo” advisor, as this technology can actually act more like a true automated financial assistant. For example, Google’s Bard generative AI assistant can address relatively niche topics, like helping San Francisco residents with home shopping or providing cross-border tax advice. And generative AI technology will only continue to improve over time.
The transition to more powerful generative AI assistants will profoundly alter the financial services industry. Over time, consumers will likely interact with their generative AI financial assistant as the first “port of call” instead of navigating around the website or mobile app. Generative AI will be able to handle straightforward questions and will provide high-level guidance, leaving human experts to focus on more value-add activity.
So, with that background on how generative AI will change the industry, the question becomes “when.” When will financial services websites and mobile apps start offering generative AI assistants to clients?
Broadly speaking, the banking, credit card, and self-directed brokerage industries are more responsive to competitive pressure and new technology. These financial services verticals tend to react relatively quickly to disruption – such as the rise of zero commission trading or innovative credit card perks that capture market share.
Perhaps unsurprisingly, the two most prominent examples of live client-facing generative AI assistants in financial services come from these three financial services verticals. American neobroker Public.com (Alpha) and Dutch Neobank Bunq (Finn) have both launched generative AI assistants. Public’s Alpha assistant can notably generate impressive commentary on stocks and the market. Bunq’s Finn assistant can perform detailed analysis of the client’s account, including granularity to the level of how much money the client spent on pizza in the prior year.
Wells Fargo (Fargo) and American Fintech Dave (DaveGPT) have also announced that they have developed a virtual assistant using generative AI, although in my testing these two assistants are fairly basic and can primarily only handle simple customer service-related questions. Presumably, the functionality for both assistants will become more advanced over the course of 2024.
Expect more bank, brokerage and card firms to launch client-facing generative AI assistants in 2024. By the end of the year, these sectors will go from a handful of examples to more widespread adoption, creating strong competitive pressure for laggards to respond with their own generative AI assistant.
According to Jason Mikula, Publisher of Fintech Business Weekly, “the excitement about generative AI-powered tools is evident, both in the number of financial services firms discussing the capabilities in earnings calls and launching beta features, but also in bottoms-up adoption from end consumers. ChatGPT attracted over one million users in just five days after the public launch. For financial services companies, there are, as always, opportunities, risks, and barriers. In particular, firms must be careful about how their assistant will handle questions seeking advice and recommendations.”
To broadly generalize, the insurance, workplace retirement plan, and traditional financial advisor industries do not respond to major technological shifts quickly. All three of these verticals typically involve strong personal relationships and/or very slow sales cycles, so there is less competitive pressure to respond to the latest technological innovation.
That said, each of these financial verticals would benefit greatly from generative AI assistants, as service and support is a significant cost for each of these products. A generative AI assistant that can hold a conversation with clients and can provide high-level guidance would reduce the routine servicing burden on insurance agents, financial advisors, and plan administrators.
Overall, expect these three financial services verticals to move slower. While there may be early adopter exceptions, it will likely be a few years before the average insurance and workplace retirement plan website and mobile app includes a powerful generative AI assistant.
Generative AI assistants are coming to financial services. It is a matter of when, not “if,” and 2024 is shaping up to be the year generative AI arrives in financial services. For financial services firms that have not started the process of developing a client-facing generative AI assistant, you must start the process now.
It may take your firm many months to align all stakeholders on the design and approach before a single line of code can be written. Therefore, it is better to start this process now than in reaction to competitive pressure later in the year. In order to help firms on this journey, next week I will publish a follow-up article that outlines how financial services firms can build a generative AI assistant.