Tokenization and Stablecoin Infrastructure: How Banks Can Unlock New Value in Payments and Asset Management

Tokenization is moving from experimentation to execution, reshaping how financial institutions issue, manage, and transfer real-world assets. This session explores how banks and payment networks are approaching tokenization and stablecoin settlement, from foundational infrastructure decisions to scalable, production-ready use cases. Viewers will gain insight into on-chain settlement, programmable payments, and the operational considerations that institutions must address to integrate tokenized assets into core financial systems.  

Transcription: 
Transcripts are generated using a combination of speech recognition software and human transcribers, and may contain errors. Please check the corresponding audio for the authoritative record. 

Neil Chopra (00:13):
Hi, good afternoon everyone, and thank you very much for joining us today. I'm really, really excited for this session and thank you to American Banker for having us. My name is Neil Chopra. I lead strategy and solutions at Fireblocks. I've been in the enterprise blockchain space for over eight years now, and prior to that, I spent 10 years in corporate treasury. Needless to say, I've been waiting for the inflection point that has been 2025 for a while now, which I feel is a critical turning point for the blockchain industry at large for a number of reasons and why we're here today. We don't need to do an assessment of what's happened over the past four, five, or six years. I think what we really want to do is level set on what we've seen happening over the past year—some of the moves that innovators and first movers in this space like Visa and Cross River Bank have made.

(01:17):
I really think in terms of ensuring that the broader market truly does understand what is going on here, we've finally shifted this narrative away from crypto as an asset class to blockchain as a technology. Now we truly are seeing utility and scaling within the blockchain industry, specifically around stablecoins and payments, proving the business value out there in the market. To that end, I'm really excited to have two of the brightest minds in this space, Cuy and Luca, who are not only helping to drive but really shape the adoption of stablecoins and digital assets as we bridge traditional markets with blockchain-based markets. I'd like to give Cuy and Luca both an opportunity to do a quick introduction and give a high-level overview of what your respective teams are working on. We'll start with Cuy and then hand it over to Luca.

Cuy Sheffield (02:22):
Yeah, great to be here. Thanks for the opportunity. I'm Cuy Sheffield, Visa's global head of crypto. We started the crypto team at Visa back in 2019, so we've been in the space for over six years now. Our approach has been that we see Visa playing an important role in bridging the existing payments ecosystem with blockchain technology and stablecoins. We've started—and we'll talk a lot more about—the work we're doing with stablecoin-linked cards, where we have over a hundred card programs that are live spending from a balance of stablecoins. We've been using stablecoins inside of Visa to improve settlement and treasury operations. We've been integrating them into our cross-border money movement business and building new infrastructure for banks to help them come on-chain. So I'm really excited for the discussion today. Neil and Luca are both fantastic builders and advocates in the space.

Luca Cosentino (03:25):
Same here. Thanks for having me. My name is Luca. I run the crypto division at Cross River Bank. Cross River is a partner bank that offers everything a bank offers but as a service for Fintech builders. We power most of the relevant Fintech programs you can find out there as well as some of the newest use cases. Relevant to this topic, we launched a stablecoin payment offering just a couple of weeks ago. We're very excited about all the work we have done, but even more so about what we can do in the next year or two.

Neil Chopra (04:07):
Awesome. Thank you both. Again, I'm really excited for this conversation. We've got about 40 minutes. We could probably talk for hours, but I know there's a lot of folks out there who may have specific questions or want to dive a little deeper into things we touch on. Please feel free to use the Q&A throughout the session. We will look to incorporate that into the conversation because we want this to be meaningful, educational, and helpful for everyone in the audience. To that end, a lot of the questions or conversations I've had over my five years at Fireblocks with traditional institutions—not just in the US, but globally—revolve around the starting point. What is this technology? How do I even go about getting started within my institution to assess where, how, and why this should fit in?

(05:06):
Given that you guys have really been driving these internally, maybe go back to "day one" of wanting to do this and figuring out how. Cuy, you guys have had a really interesting journey over the last few years. Luca, I know it's been a little stop-and-start on the Cross River side, but given that you've had to deal with internal hurdles, regulators, and understanding how this technology embeds, what advice would you give to anyone saying, "I want to start an initiative internally, but I don't really know where to get started"? Cuy, I'll start with you.

Cuy Sheffield (05:49):
It's a great question. I'll start with a bit more about our journey. We recognized that stablecoins are a technology and, even as early as 2019 or 2020, we were seeing bottoms-up adoption and demand for them in the market. Visa's been in this interesting space because we work with some of the oldest, most conservative financial institutions in the world and some of the fastest-moving early-stage Fintechs and crypto companies. We were responding to our clients where we had crypto companies looking to issue or accept cards. As we got to know them, we found they were increasingly running their businesses on stablecoins. They were using stablecoins in their corporate treasury and wanted to offer products to end customers who had stablecoins.

(06:42):
We saw it coming up more and more. Our approach even at that stage was that there's some probability stablecoins are going to cross this chasm from crypto into more mainstream payment flows. We recognized the regulatory environment was unclear and there were barriers to overcome, but we saw a scenario where they would cross from solving problems for crypto companies into mainstream finance. Our approach was to find a customer and a specific use case where we could add value for them with a stablecoin. I think that's one of the best places to start. It's so much easier inside a large organization to innovate with your customers. If you're innovating on your own in a vacuum without connection to the core business, it's very hard to prioritize that versus saying, "We have a customer, here's what they're doing today, and here's what they'd like to do with us."

(07:52):
In our case, that customer was actually Crypto.com. They were an issuer of a crypto card program operating using stablecoins on the backend. When it came time to settle with Visa, they were having challenges converting stablecoins into fiat in order to settle in traditional fiat. We said, "What if Visa could meet a client where they are and just accept the stablecoin?" If we can accept the stablecoin, it's easier for them to operate and scale the program. Then we went down the path of figuring out how to make our first on-chain transaction—which is a great milestone for any institution. When is the first time you're actually going to send or receive value on-chain?

(08:42):
We scoped it and put guardrails in place: Which stablecoin do we want to use? Let's evaluate existing ones. Who's going to be the infrastructure provider? What account do we need to receive the stablecoin to? What blockchain do we want this to happen on? How does this connect to our treasury systems? How do we manage compliance in screening addresses? How do we inform our regulators? How do we set a cap on the amount of exposure we want to have? We started this with a very early pilot and literally did one transaction: Crypto.com sent us USDC and we sent it back to them. That was the first on-chain transaction Visa ever did, and that was in 2021.

(09:26):
We were very methodical over the next one to two years to go from one transaction to a pilot where we're sending and receiving stablecoins every day. You have to find a customer who is interested, work with them, and put guardrails in place to get you to that first transaction moment. That's a major step every institution should have. If you can make a single transaction on-chain, you can then build into ongoing transactions and scale them.

Neil Chopra (10:07):
Something I've seen at Visa that's probably more established than at other institutions is that you have a lot of alignment between the crypto team and the rest of the organization. Identifying potential customers, given that you work with everybody, is probably relatively easier. A lot of what I hear from banks, though, is that they don't have the customer demand. Looking back at 2021—and Luca, this might be a great time to share the Cross River journey—banks were saying they saw deposit outflows to Coinbase and Robinhood and wanted to enable crypto trading. Most banks don't offer trading or brokerage, so standing up that business on brand new technology was a massive lift. Now we're seeing stablecoins, payments, and deposits, which are the core and backbone of banking. Luca, how do you look at identifying customer demand before it's too late, or how can you be more coordinated internally to identify product-market fit?

Luca Cosentino (11:53):
We operate with a mentality that is a little bit different from traditional banks. We don't necessarily start from our internal problem; we start with the external problem. We learned from all the innovation in financial services over the last 10 years that if we don't figure something out, someone else will—another bank, a Fintech, or literally anyone. We believe that at the end of the day, customers look for value. The question is, how do we provide it? In the last five years, a lot of value has been created around enabling use cases that were not previously possible.

(12:43):
One obvious example is instant payments. The complexity lies in shifting internal operations—technology, people, and processes—from a world where everything runs in batch to a world where everything is instant. That shift itself required a culture and process shift inside the company to allow this new world to exist at scale. We think about transmitting value to the end customer and solving their problems. The next step to that has been stablecoins.

(13:37):
I echo a lot of what Cuy talked about regarding the regulatory path to going live. Start from an existing need, take a little bit of a leap of faith in believing how big this is going to be, but do everything you can to understand the value for that limited use case. Excite yourself first and then others with what this could become. We faced trade-offs between over-investing and under-investing. By doing the work, we learned so much that we were rarely at risk of over-investing because as soon as you get deeper, something either stops you or makes you keep going. The investment is more of a day-after-day process than what you put on a slide at time zero.

(14:30):
Regarding how to get started, right after identifying the customer need, you must adopt the right mindset toward innovation. It sounds like a cliché, but it's super important. You have to look at a problem with the intention of finding solutions rather than finding more problems. In banking, that translates into risk management, which we distinguish from risk avoidance. Risk management means doing the homework, testing that one transaction, seeing what happens, going back to the drawing board, and improving.

(15:23):
After doing this a few times, you can do a hundred transactions. That initial step is what gets us going. Stay close to the customer, adopt the right mindset, do the homework, and test.

Cuy Sheffield (15:55):
Risk management versus risk avoidance is a great framework. It's much harder to do risk management than to just say we're not going to do anything. I'd add that individuals driving this inside an organization have to use the technology themselves personally. I say this all the time inside Visa: if you've never made a stablecoin transaction before in your life, you don't have a say on what the stablecoin strategy should be. It's really hard to predict a new technology and think about risks versus opportunities if you've never actually gone on-chain.

(16:52):
Imagine someone saying, "Here's our internet strategy," but they've never gone online or opened a web browser. Getting your first on-chain transaction inside an institution takes time and has to be done the right way. But every individual tasked with driving a digital asset strategy can make their first on-chain transaction in the next hour. You download a wallet and send some money back and forth.

(17:45):
We've done these sessions with our executive team where everyone downloads a wallet. We had a call with regional presidents in 15 countries and just started sending stablecoins back and forth between each other. That moment of understanding—"How does this work? I copy and paste an address. What is a gas fee?"—allows you to then start talking about institutional infrastructure. I see people talking about the space too often without actually using the technology in their personal lives to get a ground sense of how it works.

Neil Chopra (18:16):
That's a great point. Luca, you mentioned that outside of stablecoins and payments, we're seeing broader adoption of the technology. As we move toward getting the Clarity Act through in Q1 of next year, Bitcoin is becoming a financial product. We're seeing tokenized money market funds and structured products come on-chain being used as collateral for derivatives trading because it allows for 24/7 real-time movement. Embracing this as a technological evolution is critical. I'd love to get your views on the balance between finding product-market fit for customers versus the "get a wallet and understand it" approach.

(19:28):
For a bank, the individual step is getting access to a wallet technology and infrastructure, plugging that in, and understanding how it interacts with your core ledger. Understand what data you need to send to initiate a blockchain transaction. Who are the screening and Travel Rule partners you need? Those are the building blocks. I get a little nervous that we're getting to a point of no return—not that people will be too late, but the first wave of scaling is going to start happening next year. How should these guys balance long-term assessment with the need for immediate ROI?

Luca Cosentino (20:36):
We're fortunate today. Three or four years ago, answering these questions was done in isolation with a couple of forward-looking lawyers and no real system around it. Today, the banking system is a target of infrastructure providers that have done a lot of homework to understand banking before selling to banks. It's possible to launch something within a decent timeframe.

(21:29):
Regarding balancing ROI with investment, we measure ROI at a couple of checkpoints. The first is the initial one: how much do I spend to enable a use case that may or may not take off? When you look at this as a portfolio strategy of different investments, what is the return on that portfolio? It's impossible to bet that stablecoins will definitely be the dominating payment rail of the next 20 years. Once you approach it as an MVP to start getting into the reality of the tool, it becomes much more interesting.

(22:21):
The second checkpoint is seeing the benefits in what you launched. Can we have a reasonable conviction that over the next 10 years more payments will move to being instant? Probably yes. Even if stablecoins disappeared tomorrow, certain capabilities—like being 24/7, doing advanced compliance, and integrating the full stack—will stay with us as an institution.

(23:14):
You're trying to stay close to customers and get first in line for something beneficial. The third checkpoint is when you are at scale and look at unit economics rather than just revenue. It's easy to make revenue with legacy payments but hard to make a profit because of all the operational costs you need to absorb. Stablecoins may have a higher initial investment but a lower operating cost at scale.

Cuy Sheffield (24:13):
In talking to many banks, I've observed a challenging question: what are the use cases? You could spend a year just investigating and doing presentations on use cases. There's a rational approach that says you start with the use case and, once you have 100% conviction it has high ROI, you figure out the infrastructure. In this world, it makes more sense to reverse it. It's really hard to predict every end use case for a general-purpose technology like a blockchain or a stablecoin.

(25:11):
If you're in it every day, you see things you would never have predicted. Ask yourself what the lowest common denominator is. Regardless of the use case, the infrastructure is straightforward. Your customer will likely need a way to store, send, and receive a tokenized asset. A stablecoin shares properties with other tokenized assets. You can take the approach of saying, "I don't know what all the use cases will be, but I'll bet that in 2030 blockchains and tokenized assets will still exist."

(26:16):
Do the work now to set up the basic infrastructure and integrate it into your existing operations, processes, and core ledger. That's the hard work. Once you have that groundwork, it's hard to imagine a successful scenario where it's not important to send, receive, store, and convert digital assets securely. You can make the investment to light up the infrastructure and, in parallel, talk to clients to figure out which use cases to deploy it on. If you wait until you find the perfect thing, it might be too late because the infrastructure takes one to two years to get ready.

Neil Chopra (27:15):
We're seeing increasingly that banks acknowledging blockchain is here to stay aren't necessarily starting with stablecoin payments, but with the concept of a tokenized deposit. This gets you wallets tied to each customer and allows you to run existing processes in parallel. You're representing a liability on the balance sheet as a token. How do you view this concept of interoperability, and what is the difference between a stablecoin and a tokenized deposit in the short and long term?

Luca Cosentino (28:22):
Stablecoins are money put in a reserve account with a token issued against it. The vast majority of reserves—probably 85%—go into money market funds, while 15% stay as liquidity in a segregated account. One argument is that banks will lose deposits to stablecoins because what was previously a liability on the balance sheet is now a money market fund somewhere else. One answer was to create a tokenized deposit that mirrors the bank account balance and stays as a liability on the balance sheet.

(29:41):
The trade-off is that a stablecoin can travel freely like cash, whereas a tokenized deposit usually needs to stay within the confines of banks or financial institutions that handle KYC. Regarding interoperability between tokenized assets and the fiat banking system, I thought that would be one of the top reasons Cross River should play in this space. Both Fintech builders and end users are currently faced with a trade-off between how much money to keep in a bank account versus a crypto wallet. Drawing a line between the two doesn't benefit anyone.

(30:41):
We thought a good place to start bringing benefits would be making the two interoperable. Our core banking system, built from the ground up in-house, can process both fiat and stablecoins at the same account level. We enabled a platform where the account is at the center. If you want to move money through legacy rails, instant rails, or stablecoins, you can do that and everything settles in the same bank account. This helps Fintech companies build products much easier and removes friction from the back end. I believe this is going to be the norm in a few years.

Neil Chopra (31:46):
Luca, there is a question in the chat that piggybacks off what you mentioned: How important is client engagement, feedback, and product use case evolution in parallel with technical capabilities? When you have one account where different assets can interact, how do you take feedback to ensure end users can interact with it in a way that makes sense?

Luca Cosentino (32:30):
The feedback comes hand-in-hand. When we deal with Fintech companies, they tell us what their users want to enable. In very few cases are the needs completely aspirational. In addition to technical work, there is a lot of work that has nothing to do with engineering and everything to do with risk management and processes. We've done a good job of splitting responsibilities across people, processes, and technology.

(33:20):
On the people side, having intrinsically motivated staff who know what they're talking about by doing it has been a requirement from the beginning. For processes, we mirror rather than duplicate the processes used in the bank today. We thought of stablecoins as a rail to be added to the existing set rather than segregated into a different system.

(34:14):
The work we've done with Fireblocks is to reconcile accounts to the core banking system in real-time, not through file exchanges. All operational aspects go through the same controls as traditional rails. Compliance is done in the same manner and all post-transaction activity merges the on-chain and off-chain transaction. That is a hard part of what we did, but it's the right mentality.

Cuy Sheffield (35:08):
Interoperability between deposits and stablecoins is going to be really important so that you can move from traditional fiat in a bank account to a version of fiat in a stablecoin on-chain. Is a tokenized deposit a critical piece for that? It depends on your starting point. If you have a core ledger built from the ground up that is modern and cloud-based, you could do real-time interoperability between stablecoins and traditional fiat without having to tokenize the fiat. But if you're using a ledger that is decades old and only works in batch, maybe tokenized deposits are a way to have a more modern ledger for deposits to sit on that are then more interoperable. Luca, do you see it as just another ledger solution or are there things beyond that for a bank?

Luca Cosentino (36:25):
Stablecoins and tokenized deposits are very similar from a technology and governance perspective. However, tokenized deposits require the creation of a network of people who will exchange them, which will take longer than stablecoins where there is an existing market and demand. There doesn't seem to be much we can do with tokenized deposits that we cannot do with stablecoins. Stablecoins seem like an obvious step one. If tokenized deposits become the norm, it would be very easy for us to pick up that technology with what we've already built. The opposite would not necessarily be true.

Neil Chopra (37:31):
Luca, you guys had a stablecoin initiative and built for existing demand using a modern core banking system. For you, a tokenized deposit is probably not incremental enough. But given that core banking infrastructure in the US is mostly decades old, having a tokenized deposit as the starting point that is interoperable with stablecoins gives you that infrastructure to build off of.

(38:54):
We've done pilots on bond tokenization for fund managers and large banks who want to run through the process of tokenizing a structured product and compare it to the traditional world. It's a step function: how do we get this infrastructure into the bank and apply it to something we already know? Then we can start to build new products on top of that. That's the journey for the next 20% to 30% of banks that need to identify a starting point. A tokenized deposit interoperable with stablecoins is a viable one because it will take a year and a half to launch.

(40:35):
Tokenized deposits used to be synonymous with permissioned blockchains where you had to figure out how to create the network itself. Now, some of the largest projects like JPMD are being issued on the Base blockchain. It is a token on a public blockchain with guardrails. If you can figure out how to support USDC on Base, JPMD looks very similar technically, and interoperability between the two might be easier. I'm interested to see if the line between tokenized deposits and stablecoins starts to get thinner.

Luca Cosentino (41:56):
Blockchain is here to stay because these tokens are entities that contain information and travel on standards that are the same for everyone, as opposed to legacy fragmented ledgers that couldn't talk to each other. 20 years from now, we're going to see blockchains exchanging some form of value. You'll always need a wallet to receive it, a capability to send it, something to reconcile it, and blockchain analytics for compliance.

Neil Chopra (43:00):
One last question: How do you see the evolution of stablecoins, tokenized deposits, and tokenized money market funds? Stablecoins eat into the traditional business of a bank. How do you see these products coming online over the next few years to provide incremental value for customers?

Cuy Sheffield (44:01):
There's a lot of concern around the yield question for stablecoins and the prohibition of yield in the Lummis-Gillibrand bill. We've had many conversations with banks, and there's a lot of concern there. When you look at the whole family of different tokenized assets, it's hard to imagine it won't get easier for any sophisticated corporate or small business to maximize the yield they can earn on every dollar.

(45:22):
Any amount of money sitting in a deposit account for an extended period that is not generating significant yield for the end holder is going to be at risk as money moves more seamlessly. I would expect that anytime a stablecoin is at rest, the market reaction will be to auto-convert it into a tokenized money market fund. It's not that closing a loophole will mean all deposits stay in bank accounts; there will just be different market reactions.

(46:09):
As the velocity of money increases and settlement time decreases, tokenized money market funds will likely be a big part of the reserve backing of stablecoins. You'll see reserves move from off-chain treasuries to on-chain treasuries to make it easier to move between them. Consumers and businesses are going to want to earn the maximum yield possible given constrained risk posture and guardrails. You have to participate and build toward that rather than having a purely defensive mindset.

Luca Cosentino (48:00):
It's a bit of a misconception that all the money in a bank is a deposit today. Money already has different velocities assigned in liquidity stress testing. That concept is being accelerated by Fintechs and wallets. I think the velocity of money will continue to increase. Because you'll have a faster ability to turn something into something else, I wonder if more money will actually end up going into more stable places by nature of this velocity.

Neil Chopra (49:05):
Exciting times ahead. Banks are going to have to fundamentally look at business models over the next 10 to 20 years to figure out where revenue will be driven from as technology disrupts. Thank you to everyone in the audience and American Banker for giving us this venue. Please feel free to reach out to us. We're hoping for a productive 2026. Have a great holiday season. Thank you.