​

Hi all, Julie here.

If you’ve been reading my newsletter for awhile, you’ve heard over and over again that funding for our space has been pretty wild lately. CB Insights recently published a report that shows fintech companies raised $22.8B in the first three months of the year, shattering records.

Obviously, Fintech is a massive space with some segments of it garnering more interest than others. The most popular appear to be payments, digital banking, lending, and wealth management. Something that ties all of these together is how artificial intelligence can have a big impact on cash flow and speed. Think algorithmic trading, fraud detection, and portfolio optimization. Traditional financial institutions use it as well, but it tends to be geared more towards sales rather than portfolio management.

A company that I had no idea was working on this until recently when the team reached out to do a deep dive with me is NVIDIA. My sister who loves playing video games is probably far more familiar with the company than I was before writing this.

But I’m glad I got to do this sponsored deep dive since NVIDIA is starting to do some pretty cool things in our space and wants to use its tech expertise and hardware to start helping our industry with all of those AI needs I mentioned above (not to mention seeing record profits thanks in part to data center demand from industries like financial services.). While I got paid to dive into this business, I have said no to deep dives before and I don’t work with companies that I don’t truly find interesting and valuable to my readers.

It seems like a good move on NVIDIA’s part. According to its recent survey on AI in financial services, basically all companies in our space (whether startups or incumbents) recognize the key role AI is playing in maintaining a competitive advantage. 83% of respondents stated that AI is important to their company's future success, and more than ⅓ said AI will increase their company's annual revenue by 20% or more through automation.

So in partnership with NVIDIA, I’m diving into AI and how it will or will not impact the future of our industry.

The Case of NerdWallet

Let’s start by looking at one of the early fintech startups (which also is rumored to be going public soon): NerdWallet. Previously, it used information its members gave it to match them with financial instruments that had a higher likelihood for approval. These models get better over time as they learn how certain profile features such as credit scores, outstanding balances, and credit utilization lead to a customer getting approved or denied.

With over 100M unique visitors each year, it’s safe to say NerdWallet has a lot of data. It feeds that data into models, and as those models become more familiar with underwriting, they improve their ability to recommend the right personalized products.

The more NerdWallet can learn from even the most casual site visit — whether a person reads a blog, watches a video, or performs another action — the more likely it can make useful recommendations quickly. Behavioral data is combined with the hard data members provide across about a dozen different financial verticals, such as mortgages and insurance. Credit matchmaking! If you want to read more about this, check out NVIDIA’s blog on this partnership.

Transforming the Delivery of Financial Services

AI is accelerating traditional banking services that are computationally intensive, enabling companies to rapidly analyze large datasets and deliver results in real-time. This is SUPER important in financial services. It’s a big part of how payments move quickly without a ton of fraud happening, it’s how you can get approved or declined for credit in seconds online rather than filling out paperwork in person, it’s how companies verify you are who you say you are.

NVIDIA highlights Enigma as well, another fintech company that uses NVIDIA for AI in order to refine a database of about 30M small and medium businesses that their customers can access online. Enigma’s work has attracted multiple Fortune 500 clients, including financial giants American Express and PayPal. AI can also help dramatically lower the number of errors that can occur if a human is pouring over long financial docs. Virtual AI assistants (which we’ve used at Fintech Today before) are freeing up employees from tedious tasks to much more valuable assignments.

With trust being critical to long term success, finance companies big and small are using AI to enhance the transparency of their services. As an example, Fiddler Labs offers an explainable AI platform that enables companies to explain, monitor and analyze their AI products. This means that banks, for instance, can provide reasons to customers for a loan’s rejection based on data points fed into models.

Loan processing and fraud detection were the early use cases for AI in financial services. With the help of AI, companies are getting model training, accuracy, and reduced latency, all of which enhance the customer experience.

A 24/7 Customer Service Agent

That brings me to a huge part of the customer experience: customer service! Products and services powered by AI help fintech startups create engaging and personalized experiences with their customers. For example, Square published a paper on techniques for creating virtual AI assistants that are sympathetic listeners. “Square Assistant” has expanded from a virtual scheduler to a conversational AI engine driving all the company’s products.

Recent data showed that Square Assistant could understand and provide help for at least 75% of customer’s questions, and reduced appointment no-shows by 10%. As Square Assistant expands from dozens to thousands of use cases, its mix of small, medium and large Natural Language Processing models expands and requires more training. The performance improvements here can be huge. For instance, NVIDIA says that Square experienced a 10X speed up in model training through using NVIDIA GPUs.

The Future of Banking is AI-Enabled

With AI creating new opportunities across personalized recommendations, fraud detection, and customer service, NVIDIA is betting that it can help companies leverage this to reduce wasted time and money. It argues that with its GPU-accelerated machine learning and deep learning platforms, data scientists across the financial services industry can deliver results in days.. Ultimately, Fintechs are working to AI-enable hundreds of applications that will increase revenues, decrease costs and improve customer experience so they need an accelerated computing platform that scales to the demands of the AI powered bank. You can learn more aboutNVIDIA’s solutions for financial services here. And if you’ve used any of these products before, from NVIDIA or its competitors, I’d love to hear your thoughts! Simply reply to this email and it goes straight to me :)

​