Ep 48: The Future of AI Banking with Ally Chief Information, Data, & Digital Officer Sathish Muthukrishnan

‍Evan: Hi there, and welcome to Enterprise Software Innovators, a show where top tech executives share how they innovate at scale. In each episode, enterprise CIOs share how they've applied exciting new technologies, and what they've learned along the way. I'm Evan Reiser, the CEO and founder of Abnormal Security.

‍Saam: And I'm Saam Motamedi, a general partner at Greylock Partners.

Today on the show, we’re bringing you a conversation from August 2024 with Sathish Muthukrishnan, Chief Information, Data, and Digital Officer at Ally Financial. Ally is an American bank holding company, with over $8 billion in annual revenue and over 11,000 employees. The company provides auto financing, mortgage loans, banking, and other financial services to 11 million customers across the United States.

In this conversation, Sathish explores Ally’s groundbreaking journey from mainframes to AI-powered banking, the game-changing role of generative AI in customer support, and his predictions for how AI will redefine personal banking in the years to come.

‍Evan: Thank you so much for joining us. Appreciate making the time and, um, to come share some wisdom and experience with our audience. Maybe to start off, do you mind sharing or giving our audience a little bit of background about your career and maybe how you ended up in your role today at Ally?

Sathish: Well, I started off as a hands on mainframe assembly language programmer, if you will. Started off with Singapore Airlines. Always wanted to come here and work in the U. S. So, when United Airlines came looking for mainframe engineers,like me, I moved to Denver, Colorado, worked for United for a little bit. And then moved to American Express in Phoenix, Arizona. Then worked at Honeywell for a little bit. And now, for the last four and a half years, I've been with Ally and loving every minute of it. 

Evan: That's awesome. And maybe just for context, for audience, right, that may not really kind of fully appreciate the scope and, you know, importance of Ally, do you mind sharing a little bit kind of what you guys do and how you help your customers and clients?

Sathish: Ally is a full service digital bank. We are the largest direct to consumer bank in the U. S. We are one of the largest auto lenders in the nation. We have relationship with over 21, 000 dealers and produce over 40 billion dollars in auto loans every year.

We love serving our customers, and my job entails all the way from product, UI, UX, technology, data, digital, security, network, and operations, into an accountability for my team serving one of the premier, digitally native organizations in the U. S. 

Saam: Maybe let's zoom up from Ally and just talk about sort of what's happening in technology more broadly. And I think it's hard, we're talking about this before we started recording, to have a conversation in 2024 without, uh, without talking about AI. 

So Sathish, like, through your career, you've been a part of many technology waves and transitions, just starting at a very high level. Like, what's your read of this one and where are we in the so called hype cycle as it relates to AI? 

Sathish: Yeah, I mean, I think it's gone beyond the hype cycle, because it has been around for a while and now AI is having its moment. And AI is having its moment because of the convergence of other hype cycles that has happened. Whether it is data, whether it is cloud, whether it is app, native app development, explosion of technology investments.

And then the fascination towards SaaS, you know, think of a time when you had to develop everything from ground up. So the progress the technology can make as a team was limited to how many people you had and how quickly they developed. We talked about fintechs and startups and providing software as a service. So, you then weigh, how long does it take for me to develop this feature versus using something that is available, and can I enhance it?

So you're already at an advantage in terms of creating experiences for your customers. And then, if you rightfully went through the data journey, and you have cloud enabled, and you've made it findable, and you've made it easy for us to compute the data, then you have your application sitting on cloud. that it has the ability to now access the data and use the insights, then you can take advantage of this AI explosion, which luckily for us, we invested in our data journey. 

Our 90 percent of our data is in a single warehouse is on cloud. 

Saam: Wow. 

Sathish: We invested in cloud intelligently. 76 percent of our applications are cloud enabled, and we actually rebuild our network to be software defined, which allows us to use that data efficiently for the applications that need it. And then AI, Gen AI happened. So it was a perfect inflection of intelligent and responsible investment in technology, meeting the largest technology innovation that is reborn right now. AI has existed for decades now. Now that it's reborn, we are super excited to take advantage of our investment and see what AI can help enable us.

Saam: Yeah, I love the way you frame that situation. Like one of the things Evan and I talk about is like, you know, many people say like, AI is the next big technology wave, right? It's like the transition from on-prem to cloud or, or, uh, desktop computing to mobile. But I think one thing people miss is this wave is building on those other waves and it's sort of like compounding. And so like the impact is even larger.

And it kind of connects back to what you said, which is like, there's all this data investment you made and cloud investment you made that now enables you to actually take advantage of AI. And candidly, I think when I talked to some other technology leaders who haven't made those investments, they like want to jump to the AI project and then they realize they have a huge multi year data project to go get done.

You have such an interesting vantage points as you guys have been ahead on a lot of these cycles. Like, what are the two or three most interesting ways that you're using generative AI today at Ally? 

Sathish: You said it extremely well, Saam, that this technology is building on the other evolutions that has happened. One of the things that we also did was, Because technology is evolving, can we dream and be courageous enough to have a single digital canvas for all of Ally businesses? We call it OneAlly, and we have now a single app to serve all the businesses and all the customers that are interacting with us. We have a single digital landscape for all of these businesses. 

Now, what the team did behind the scenes to get that, get us to that point is significant. We no longer talk in terms of iOS and Android development. We talk in terms of React Native deploying code that will run on both platforms. We talk in terms of Fondly, we call it Metronome, a single framework that allows us to build a web development that cuts across all businesses and consumers. So we arrived at that through a journey we call OneAlly. 

Now that you arrived at OneAlly and you have invested in all of this technology, what is the next evolution for us? We want to go from a product centric orientation to an experience centric orientation. We believe the customers of tomorrow and a large portion of customers of today are driven by experiences. They are no longer consuming products, they're looking at what experience is going to fulfill my life. What experience is going to make my life better? That is why travel is taking off like crazy right now. People are not buying products, right? They want experiences. So we said, Gen AI could help enable that. And, and let me give you one simple example.

Customer care associates talk to our customers on an average 10, 000 times a day. Every call is about 12 to 15 minutes. Again, we are focused on customers. We are focused on solving their problems. In the past, they have to keep track of the conversation and summarize the call at the end of every call. And it takes about, takes a few minutes for them to do it. But humanly, they have to remember what happened during the 15 minutes. Enter Gen AI, it's capturing the entire conversation, transcribing it, and creating a summary for our associates to look at it and say, these are the conversation, this is the conversation that I had. Here is the takeaways and just click approve. So one example of how Gen AI has significantly changed internal customer experience for us. 

Saam: So maybe I have two questions. And with the lens of like, I think there are other technology leaders listening and, and trying to think through like, how can I have the same impact with AI that Sathish has been able to have at Ally. And so maybe two questions. 

One is like, how do you actually choose what use cases to go focus on? Like, there's so many things you could go do with generative AI. Like, what was the framework or approach that led you to choose, you know, customer support and care as a first use case? And then the second is like, once you choose, and you know, how do you think through the build versus buy? And like, there's a number of emerging vendors. There's great tooling and infrastructure. Like, what's been your philosophy there? 

Sathish: Generative AI is new and fast evolving. So with like any new technology, there is a number of risks inherently that you have to tackle. And two, we are a regulated institution. We then came up with simple guiding principles for how we would use Gen AI and it was very simple. Number one, let's build use cases that face off to our internal customers as the technology evolves. Number two, there should always be a human in the middle. And number three, I cannot allow sensitive information to leave Ally.

When I laid out those principles, My build versus buy decision became very clear. I can use the technology that is available to solve the initial set of use cases. So I don't need to go build a new LLM. But at the same time, I had to build an Ally AI platform to ensure that sensitive information doesn't leave, but also to future proof the use cases that I'm going to build.

I'm going to use the technology that is available right now, which is open AI three dot five. How do I future proof myself to use another LLM that is coming along, in the future? So we built the LIAI platform that allowed us to now interact with the first LLM, but has the ability to go integrate with any other LLMs. I can send the information to them, one or multiple, maybe consolidate the answers or pick and choose the best answer. 

And then also, include the context for our own internal customers by retaining it within Ally AI, and I'm not sending that information outside. So having those simple guiding principles allowed us to solve the build versus buy problem, but also allowed us to prioritize the use cases that fit within those principles.

Evan: If I'm an Ally customer today, help me dream about the future, right? What will Ally look like in five years? And how's that going to change the customer experience? Like what new capabilities will I get as a, you know, user of the platform or as a client or customer that, you know, maybe your average, you know, the average person familiar with Ally might not be, you know, like maybe not think of today.

Sathish: Yeah, you know, we talked about how we can, we should shift our mindset from product centric thinking to experience centric thinking. And what that entails is, to a certain extent, anticipating your customer's needs. 

I fondly explained that prior to experience shift, like our customers, are not looking for an auto loan. They are not looking to park their money. They are not looking to get a house mortgage. They're looking for transportation. They're looking for wealth creation. They're looking to build a family, create memories. How do you enable that? Can you start teaching them financial responsibility? Can you give them a platform where they understand where to save and grow their money? Can you give them a platform to understand how to invest responsibly? Can you give them a platform to say here is when you can get your house for this price, um, and not be a burden on your lifestyle? How can you help them educate? How can you help educate them financially? All of that can be enabled through terrific experiences in a way that every customer feels like they are our only customer.

Being a digital native organization, through technology and experiences, we could perhaps help enable that. And if I think about a vision of the future, that would be my vision. Every customer feels like they are our only customer and Ally is here to serve them in a way that nobody else can. 

Saam: How else do you think the future could transform with AI? And like, are there any other sort of specific ideas you have that maybe feel like science fiction today, but will feel very real five to 10 years from now? 

Sathish: Yeah. Let's, let's learn from history. Nobody thought we would jump in a stranger's car and immediately trust them. Inherently trust them. They are great drivers. They would take me to my destination. Nobody ever thought that we would live in a stranger's house. and think it is safe. Um, nobody ever thought of a time when a car can do autonomous driving as quickly as it did and as efficiently as it does. So, what the future holds is going to be what has become common knowledge now.

And even from a privacy and a security standpoint, how often would you have given your home address to a stranger? Somebody calls in and says, Hey, I'm bringing you a McDonald's burger. Give me a home address. You don't hesitate to give out your home address, which you do now, right? Nobody ever thought that finding your date of birth or home address on the internet was safe. That was five to six, 10 years ago, but now nobody even gives it a second thought. So the thinking around what technology can enable is going to significantly change. 

I could predict, perhaps, that the healthcare industry would significantly evolve. It would tell you it's constantly measuring how your internal organs are working and perhaps sending some a signal to your doctor and they are proactively reaching out to give you medication or asking you to book an appointment. Or it is booking an appointment. Um, that's feasible. 

It could potentially plan your kid's birthday party, send out invites for your kid's friends, figure out what needs to be ordered based on who said yes, figure out a place to book the party, and then do it. This is not outside the realm of possibility with the technology that we have today. But how can you bring connecting the dots, critical problem solving, and help your employees, your customers, use technology effectively and safely is where you would be differentiated. And I'm intentionally not giving any banking examples because I don't want to give away what we're working on. 

Saam: Yeah, no, that makes sense. And also I was laughing as you started because on your point around like walking into a car, uh, you know, we don't, we don't know the driver. Next time you're in San Francisco, you'll, you'll use the Waymo app if you haven't, and you walk into a car that has no driver. 

Sathish: That's exactly right. 

Saam: And I think, uh, that piece of what felt like science fiction a decade ago is very real. And by the way, it's a great product and experience here. 

Sathish: Yeah. 

Evan: It is amazing just like, you know, when you think about how fast technology has already changed, right? Like the biggest bulls of like, you know, the internet 30 years ago, underestimated the impact they now have on the world. Right. And so you're just like, you look at the spectrum of, you know, bullish and bearishness for AI today, like 30 years from now, where will we be on that spectrum? Right. 

Like, my guess, it's like above the average. Right. And it may be some scenario where we dramatically underestimate things because humans aren't good at like predicting exponential growth. 

Sathish: That's right. I would, I would tell you, maybe not focusing all on the positives on the flip side. We have to figure out how the bad actors would use AI. 

You know, our number one job is to protect data, protect our customer information. We also have to be cognizant of how. And how might the bad actors handle AI and continue to evolve our defense mechanisms? And, and that is, that is a massive challenge facing not just technology leaders, any leaders of any company across the, across the, across the globe. 

Evan: Yeah, I mean, I think it's, um, unfortunately, right. The, the history has shown that the bad guys adopt new technology faster than good guys, right? Although I think the good guys, you know, they're going to win in the long run. 

What are some of the use cases there you think that, you know, the average person in your shoes should be thinking about, right? Maybe not what's happening today, but like a year or two down the road. What are some areas where like, you know, you think, you know, some of your peers should be thinking about it, kind of preparing for it to make sure that they're not sitting here in two years with regrets about not doing something about it.

Sathish: Yeah, I would focus on ensuring that AI fatigue is not happening. 96 percent or higher security compromises have happened because a human clicks on a link through phishing. How do you make sure that you continue to raise awareness for your humans, because at the end of the day, you need to take care of the humans. AI can take care of itself. But how do you take care of your humans? You need to raise the awareness, continue to train them, continue to show them the fallacies of clicking on a link and exposing ourselves negatively. That should be a major, major focus. 

And then the second thing that I would say is depth and defense is extremely important. It is not a single layer. It's not a one and done. You need to see how you can protect yourself across multiple layers of where the data is stored, how it's retrieved, where it travels and where it lands. You need to constantly think about how do you protect yourself across multiple layers of this.

And then number three is, do not forget to experiment. We talked about, uh, analysis paralysis. That's not going to take us anywhere. Do not be afraid of the fast evolving technology. Continue to experiment, have your team be empowered. So you are making small progress. Whether it is in the offensive side or on the defensive side, it's extremely critical and important for you to be prepared, whether you did all your data journey or cloud journey. Regardless of that, you need to continue to be familiarizing yourself with new technology so you can be prepared whenever the time is, both offensively and defensively. 

Evan: I appreciate you sharing and I am going to, um, resist every entrepreneurial bone in my body to not turn what you just said into a sales pitch. And instead, we're going to go jump into our final section of this, which is a bit of a lightning round. So we'd like to kind of do a final section where get a couple of kind of, you know, quicker questions, quicker answers, right? So kind of looking for like the, the one tweet version, but, uh, Saam, you maybe want to kick it off for us? 

Saam: Yeah, absolutely. So maybe to start Sathish, um, how do you think companies should measure the success of a CIO? 

Sathish: How technology ultimately enables the businesses and the functions is how a CIO success should be measured.

Evan: What's one piece of advice you wish someone told you when you first became a CIO or technology leader? 

Sathish: You're going to make a number of decisions that could make or break your career on a daily basis. Just trust your intuition and experience. 

Saam: That's that's great advice probably for uh for many of us in many roles.

Um, Uh, maybe to switch gears to the more personal side. What's a book you've read recently that's had a big impact on you and why? 

Sathish: I loved a recent book that i've read and i'm reading again is the idea flow by Jeremy Utley Um, it You know, we talked about analysis paralysis. It gives you tools to continually think and see how you can enhance your thinking and be more innovative.

Evan: Maybe a final question here, um, what do you think will be true about technology's impact on the world that maybe most people would consider science fiction today? I'm looking for your kind of like contrarian view, right? Yeah. What do you believe that maybe other people don't, right? I want to get your unique perspective.

Sathish: Technology is going to make humans live longer. You know, there is examples of a 3D ear being replaced, um, to allow somebody to hear. You know, we know examples of what Neuralink is doing. So 40, 50 years from now, humans can decide when they want to die because technology has given them that power.

Evan: Sathish, thank you so much for taking time to join us today. I'm super excited to chat with you and looking forward to chatting again soon. 

Saam: Thanks a lot, Sathish.

Sathish: Thank you. I enjoyed our conversation.

Saam: That was Sathish Muthukrishnan, Chief Information, Data, and Digital Officer at Ally Financial. Thanks for listening to Enterprise Software Innovators. I’m Saam Motamedi, a general partner at Greylock Partners.

‍Evan: And I’m Evan Reiser, the CEO and founder of Abnormal Security. Please be sure to subscribe, so you never miss an episode. You can find more great lessons from technology leaders and other enterprise software experts at enterprisesoftware.blog.

‍Saam: This show is produced by Josh Meer. See you next time!