BNY Mellon CEO Robin Vince on Eliza, their internal AI platform powering 125+ solutions and digital employee agents

Lambda Lightning round. But let me tell you about Lambda. Lambda is the super intelligence cloud building AI supercomputers for training inference that scale from one GPU to hundreds of thousands. And without further ado, we have Robin Vince from the Bank of New York Melon. Robin, how are you doing?

What's going on?

What a wonderful backdrop. I think we are struggling.

Really pretty really remarkable.

Oh, how are you doing? Uh sorry we we we missed you on that intro because of the audio. Uh can you introduce yourself and the company I suppose?

Sure. Well, it's great to be with you. Congrats on uh and everything you've been achieving. It's been fun to watch your progress. Uh so BNY we are the world's uh real looking after the wiring and financial security of the financial system. We also are America's oldest bank. So we've been around for a while. We've seen a few cycles over time. There were just 20,000 people in New York.

Yeah. Who started the bank again?

Alexander Hamilton, the now famous dude.

That's incredible. Incredible. So uh obviously a lot of a lot has changed over the uh what 250 roughly year uh history of the of the firm. What is your day-to-day like? What is what what is on the top of your list of goals for 2026? What are the challenges to running an institution as old as BNY?

Well, we've been really reimagining the company over the course of the past three years. We've been a strong and safe part of the financial services system and the US financial system as you say for 250 years. We were the first stock ever traded on the New York Stock Exchange.

Fantastic.

Exactly. That's right. There's some there's value to longevity, but you to get old means that you learn how to adapt. And this is a world full of change. And so we've been laser focused on really bringing ourselves truly into the modern age. AI very focused on AI and everything we're doing with our platform. We can get to that. Yeah. But actually taking the company and making it be the performer that it can be. We do all these great things, these different businesses that we have. Living up to our potential has really been the motto of the past three years for us.

Yeah. as as you've witnessed AI progress over the last couple years, what do you think the biggest opportunities uh for uh for AI within uh I would say leg just legacy financial uh institutions overall. uh you know we we had someone on earlier that was talking about you know a lot of uh some of this code still running on mainframe computers and and how much uh how much modernization that I'm sure you've been working on that's not even related to AI and then how can AI kind of continue to serve that

I think for us there are three vectors and to be fair I think they're probably the same for most firms these days and I think that's true for technology software firms I I think it's true in the financial services sector as well. One is just doing what we do but just doing it in a better and more efficient way. So that could be modernizing code. You write cobalt. We have some main frames. Mainframes are actually pretty good super high volume uh use cases but we don't want them to run on old code. So having the opportunity to make that change and frankly just make it easier to be able to make the change. We've been putting it off for a while now. We've got great tech. It's cheaper and easier and quicker to be able to do it. Yeah.

Um but also reimagining processes end to end. being able to essentially take multi-step things and collapse them down to be just more efficient, quicker. So running the company better using AI is number one. Second vector of change is actually the products and services just enriching what we do and actually what we offer to clients with AI so that they perform better provide additional insights can actually be executing quicker for clients and so we just become better as an institution in our actual delivery uh of our products and services and then the third one I think is actually expanding the perimeter of the firm. We're not trying to turn ourselves into fundamentally a different company, but we are interested in taking inspiration from other things that have happened in the world like Amazon is a good example. Amazon Web Services once upon a time it just served Amazon then they externalized it as a service and it essentially became their cloud service. We've got lots of things that we do at super scale for ourselves that we could also do for our clients. We are the number one custodian in the world. We custody $60 trillion of assets. We have payment rails.

Yeah, it's it's it's fun to have scale.

You just said the biggest number ever of any company,

any CEO on the show and that's hard day moves through BNY. It's

we touch 20% of all investable assets in the world. So that is mega scale. You could sound it again if you want. I love it.

Throughut the data from that throughput allows us to be able to deploy AI into really making the company kind of different. So this third one spanning the perimeter doing for other people things that we currently can only do for ourselves is kind of an interesting thing too. Uh there was an interesting story in the Wall Street Journal today about Mark Zuckerberg looking to build a custom AI or you use AI to answer questions in his role as the CEO of Meta Platforms. And I'm interested in how you as the CEO of BNY could either are currently using AI or or think in the future you might be using AI as a co-pilot for what you do, answering questions about the business. What's on the top of your wish list if someone were to build an AI system that knew everything about your business and could help you do your job as CEO better?

Well, we have the beginning of that right now inside our company. So, we have our own AI platform that we built. We started the journey three years ago. It's called Eliza.

Yeah.

Uh homage to Alexander Hamilton again. and lies are connected to all of the LLMs but is a really a multi-agentic uh platform that are people build agents and we have live agents multi- aent solutions in production doing work today we also have digital employees that are essentially multi- aent solutions but wrapped with a user ID a login a persona and name that actually make them addressable and deployable in the company to work alongside inside our human uh employees as teammates and Eliza's an app and I do exactly what you just asked basically multiple times a day. Okay.

So I the client earlier on, hey, I'm going to be seeing this client. Could you please tell me what are the additional things that I should be talking to them about? Give me the top three ideas that are meant to be my pitch to the client. It goes through looks through every call report that we've interacted with the client. what I did last time I met with them,

what's happened in the news so far, what's pertinent and news to their business, and actually just advises me on how to be a better kind of extension to the client rep team when I'm meeting with them. So, I do that today.

Yeah. Wow.

Uh I mean, can you explain to us a little bit about what you're standing in front of? Is this some situation room uh to understand the global economy, your business particularly? It looks amazing, but I'd love to know more. So you can and you can just see a glimpse of it behind me, but yes, it's a it is exactly all of what you just said. It is a situation room. It's our cyber technology command center. It is AI powered. So AI is kind of the first line in actually looking at it, triaging it. So we're watching $3 trillion of money every day. We're watching 25 to30 trillion worth of US Treasury settlements every day. We're watching our $3 trillion worth of wealth assets and we're basically helping our clients to be able to get insights into their business that's running through our technology because we're kind of a platforms company these days. We're a bit of a non-traditional bank uh in terms of what we actually do. So, this is a virtualization. So it's all real data, real time that is actually and we have this in three places in the world that runs off private and public cloud that actually allows us to get the insight into what's happening in our business all the time real time.

I'm sure clients are asking you about disruption from artificial intelligence, geopolitical instability. What else is at the top of the minds of your clients as we go into 2026? It's a it's an uncertain time. Where are people searching for certainty right now?

Well, you know, it's a it's an important question, John. And so AI is for sure, right? You're right about the geopolitics. The world's complicated place. Digital assets is another example. Sure. Which positioned our investment in digital assets, tokenization, tokenized deposits, stable coins, tokenization of other uh security types and and frankly other financial assets or maybe welcoming new financial assets into the system. That is something that our clients are super interested in because they want us to help them manage that transition from the old or traditional rails into the new ones and we can be a partner for them with that.

What is what do you think is next? I mean it there was a lot of there was a lot of talk years ago about tokenizing assets bringing you know assets that we know and trade every day onto you know digital rails cryptocurrency onchain uh that it feels like that happened with the dollar it feels like stable coins are here they've arrived real estate has been slower stocks have been slower what do you think

picked up in the

is oil on chain now I was not familiar with that but

people are trading oil on chain but it But I'm sure I'm sure you see some some you have some questions around how that's actually being done. But

yeah, well look, remember with oil as well, it's a physical commodity at the end of the day. The whole concept on chain of the mirror world uh is like if if ever there was an example of of just a reminder that it's a physical asset today.

But look, let me give you my gen one, gen two, gen 3 view of that. And even before I do, let's just acknowledge where we are in Gen one. So tokenization, yes, we're underway. We're beginning. You mentioned stable coins. They have market share this big right now. And so it's early days and it's a little bit like AI. You can see the promise of the tech. You can see the opportunity, but at the end of the day, adoption, full integration into the system is very different than just having the tech. And we talk all the time in our company about AI. Great, great, exciting, but it's got to be deeper. It's got to be everywhere. It's got to be for everything, for everyone. So I'd say digital assets similar situation even in gen one we're beginning it's early days lot of promise opportunity

but just taking a an existing asset or even a physical asset that exists that isn't really part of the financial system putting it on be the beginning then how do we actually make things work better and differently because for some existing assets they're not necessarily better just because we tokenize them you know we take a share of Microsoft and we declare victory because it's a token We haven't really changed much in the world. If you take a bond, the question is not only can you tokenize the security, but what can you do with the indenture? How can you have the if and therefore statements that are all part of that piece of documentation and actually encode them in a chain in a way that actually is going to add additional value? That's a different thing than just tokenizing the security itself. And then Gen three, if you actually think about financial securities as ultimately being a cash flow and terms and conditions, then maybe you can just rerender financial assets in entirely different ways because of the fact that you now actually have the blockchain and you actually have capabilities that didn't exist today in the traditional rails. Now today's L1s aren't really up to that task. So there's innovation that's required at the L1 layer level as well as in the L2s and then for for participant firms. But that's how I see the world. We're early. This is a 5year 10-year journey. It'll happen. I think it'll happen more slowly than AI. But that's another example of an innovation right now. Earlier in the show, we were talking to Mitchell Green from Lead Edge Capital about private credit, and there's been some reporting around just how big that non-bank lending industry is. How should average investors who aren't maybe specifically in a private credit fund, but just, you know, your average American trying to make sure that their portfolio goes up every year, how should they think about private credit, the risk in the system, how they can protect themselves or potentially benefit? Where is America when it comes to the private credit industry right now?

So, first of all, just a reminder that private credit is first of all, credit.

Yeah. And so we've got private credit, we've got public credit, and they're essentially two different markets, different investor types potentially go in, but at the end of the day, they're trying to solve for the same thing, which is leaning in to being able to lend money into the into the financial markets and capital system, give entrepreneurs and other business builders or business runners access to more capital. And so there are some types of borrowers who are more suited to the private markets and some who are suited to the public markets and some who do both. There's a lot of drama at the moment in private credit and there are some good reasons for it because there have been some exposures that are bit being a bit idiosyncratic to private credit and also there always liquidity issues when people see headlines and they say oh my gosh is my money safe and then people their money back and then you have the old elevator elevator door problem. more people want to get out, but there's not necessarily enough space and there's a bit of a rush for the exit. So that's exactly the problem that that we're seeing right now.

But the market, I would observe, is today still pretty strong because the credit market's strong and ultimately the economy drives the credit market. Now, we have high energy prices for a long period of time or we have a real weakness in the economy, then we're going to have problems in the credit market and then we're going to have real problems in the private credit market and public credit market. So I think today we're in this spot where there's a lot of focus. It's a large but not huge market compared to the scale of the public credit market, but the underlying fundamentals are mostly pretty good, but there's a huge watch out with energy and as a result a huge watch out into the readout into the real economy. And if we have high energy prices or other cracks and they persist, then it's a different story.

Yeah. with uh the the health of the underlying economy. I feel like people track inflation, they track GDP, they track the stock market. Uh if someone's trying to, you know, now today they're an oil expert. Uh if someone's just trying to get an understanding of how the US economy is growing, what are the signals that you would recommend that they actually uh look through or try and understand? Well, there's first of all, when you're really looking to the economy, you're looking at employment is super important. People who have jobs feel more comfortable and people who feel more comfortable go about their lives and they spend and they employ back and they help the economy to grow. So, it's really important to watch uh in employment and also to watch sentiment.

Then then consumer sales and spending and kind of credit card health and are people being overleveraged in how they're going about their spending. Those things are important. uh what is the what is the profitability ultimately of companies and are they growing uh are they are they investing are people capital deploying energy is a huge input high energy prices have a very negative impact on the United States we sit here and we feel very good about the fact that we've got a lot of security and energy supply and so actually having the molecules that can flow because we have access to supply is a really big deal but we're an export market so at the end of the day the world prices sort of set the price domestically here as well. And if we have high energy prices, it ripples through into food. It ripples through all into all of the different things that are energy dependent. And these days, kind of everything is energy dependent.

Yeah. Well, I guess the good news is that oil's down 10% today. I I feel odd uh saying that, but it was it was uh very nerve-wracking when uh oil seemed to be going up and up. And as you mentioned, high energy prices can really put a drag on the economy. So hopefully I think we're all uh hoping for stability in financial markets, stability in uh energy markets and all of the above. But uh thank you so much for taking the time.

Yeah, really enjoyed your

to break it down and come back on soon.

Yeah, we'd love to talk to you more.

Have a great rest of your day.

Cheers, Robin.