GlobalFintechSeries Interview with John Sun, CEO at Spring Labs

John Sun, CEO at Spring Labs chats about Spring Lab’s new AI-powered copilot and the eventual impact of AI across fintech:

__________

Hi John, tell us about yourself and your role at Spring Labs. How has Spring Labs evolved in the last few years?

I’m the CEO at Spring Labs; we build AI technology for financial institutions. Practically speaking, most of my day-to-day job is to speak with our clients to understand their business priorities and then bring back those insights to our AI researchers to develop AI products that move the needle for our customers. We started in 2018 with a pretty simple thesis—financial institutions, and their customers, win when they can quickly incorporate the latest innovative technologies. We developed that thesis at my last company, Avant, where we were one of the first in the industry to leverage machine learning (in 2012) and create better financial products for end consumers years before it caught on with mainstream financial institutions. Reflecting on that experience, and why it took so long for the rest of the financial world to catch up, we started Spring Labs to help quickly close the “last mile” gap between cutting-edge technologies and practical adoption. Today, we’re seeing the same excitement and the same challenges with generative AI, and think there is massive potential for companies that can bring easily adopted products that solve real problems to market.

We’d love to hear more about Spring Lab’s new AI-powered copilot and how it enables end users?

Our CustomerAssist AI product helps your users navigate the information and functionality already available on your web and mobile application today. We call it a copilot instead of a chatbot because it’s really designed to empower the user directly rather than replace customer service. However, because we can help you simplify the user experience for your customers, you’ll also see a decrease in the amount of customer service contacts, which in turn reduces customer service costs while making your customers happier. An example is if you have long documents like loan contracts which are difficult for the average user to make sense of, our universal copilot experience can help those users ask direct questions and get answers. Another example is we can turn traditional web application forms into conversational applications, where the copilot asks the user for relevant information and then fills out the form on behalf of the customer.

As users move from traditional banking modes to newer fintech: how do you feel the future of fintech is set to shape up to support expectations and needs?

Banks are necessarily slow-moving, and that’s a good thing for consumers. We’ve seen in the past what happens if banks try to move too fast, and oftentimes it ends in negative consequences for the financial system. The original vision of fintechs was to replace banks, but as time went on, fintechs have evolved into much more of an interface layer between banks and consumers. This system seems to be working well, fintechs can focus on innovation while the banks behind them focus on risk mitigation and compliance. AI adoption is going to depend on banks and fintechs working in tandem to make sure that we’re bringing new experiences to the consumers, but doing it in a way that’s responsible.

Read More : Why Human Testing Is Critical for Fintech Companies in the Age of AI

Can you share your thoughts on the impact of AI on fintech and also, what are some of the most unique AI-powered fintech features from around the world that have piqued your interest?

AI represents probably the biggest shift in fintech capabilities in the last decade. In the near future AI-native experiences will start replacing existing fintech experiences just like the web and app experience has displaced traditional branch-based experiences for a lot of the financial sector. There’s been several articles in the past few days from big names like Goldman and Sequoia on the inflated, and sometimes unrealistic, expectations around AI adoption. I think what that really means is we’re still limited by our imagination on what AI can bring. The early applications of AI, like automating customer service, are still rudimentary but will lead to more impactful use cases down the road. Some of the more innovative features I’ve seen are things like using AI to calculate granular unit economics and deep analysis of consumer spending patterns. We focus heavily on using AI for regulatory compliance and CX, I think that’s a greenfield opportunity as well, since a lot of the intractable challenges in compliance are easily solved with gen AI, and compliance has always been an underinvested area within financial institutions.

What should today’s fintech leaders keep in mind as they look to power their products with better AI-powered features?

AI is just another technology at the end of the day, and the focus should be on solving existing business problems and creating better customer experiences, instead of adopting AI for the sake of adopting AI. This is pretty generic advice, but somehow every tech cycle always starts with investment in the tech for the sale of the tech rather than its benefits. Also, look for opportunities to rapidly prototype across a number of different use cases rather than invest heavily into a single implementation. Look for 2-4 week POCs and rapid ROI rather than committing upfront to annual contracts that are estimated to pay off in 3+ years.

What are some of the top myths around AI and finance that you’d like to bust here.

I’ve recently noticed a bit of focus on using gen AI for underwriting, and I don’t think it’s the best tool for the job. There’s still so much unexplored territory in traditional machine learning techniques that can help level up underwriting. Using Gen AI for underwriting presents unique challenges, like explainability and biases, which will take time, potentially many years, to work out. The one area in which gen AI can help with underwriting today is structuring unstructured data such as call logs, or agent notes, into attributes that can feed into traditional machine learning techniques, and I’m sure that area will yield fruitful results soon.

How would you describe the future of banking?

Proactive experiences. Right now banking is still very reactive, tools like overdraft protection are a bandaid, but ideally, you address challenges before they become problems. In the near future everyone might have a personal AI banker that helps proactively manage their finances, and flags the risks and consequences of actions in their day-to-day financial life. Instead of financial institutions reacting to complaints they may introduce predictive AI models that can highlight compliance issues before they come up.

In what ways can fintechs empower users with better experiences: what key factors matter to overall customer success and customer experiences in fintech?

I don’t think gen AI has changed what fundamentally makes for customer satisfaction. As fintechs are designing gen AI experiences it’s important to make sure that these experiences enhance what customers are looking for already—transparency, simplicity, empathy. Our partner program members have confirmed that most consumers prefer self-service. Gen AI copilots like the ones we are developing at Spring Labs enable more customers to solve their own problems, more often. At the end of the day, consumers want to spend as little time managing their finances as possible and gen AI is likely to help accelerate simple tasks while helping fintechs and sponsor banks to increase their compliance coverage. A true win-win. The key is partnering with the right gen AI company that understands CFPB regulations so the CX and efficiency gains are not followed by an enforcement action.

Read More : Global Fintech Series Interview with Christy Johnson, Chief Product Officer at Versapay

[To share your insights with us, please write to psen@itechseries.com ]