The Glenbrook team is taking stock in June 2026 to see how the predictions from our January episode are holding up. Tune in for updates on trending topics in payments, including agentic commerce, stablecoins, cross-border and instant payments, risk and fraud, and the regulatory landscape. We also share insights on the “hot off the presses” news on tokenized deposits and the Visa/Mastercard interchange settlement. 

 

 

 

Episode Transcript:

 

Drew Edmond: Hi, everyone, and welcome back to Payments On Fire. I’m Drew Edmond, a partner at Glenbrook, and I’m your host for this episode. Back in January, we kicked off the year with one of our Fanning the Flame segments, where we got together to share our predictions on the hot topics in payments for 2026.

Well, here we are at the halfway mark. It’s June 5th when we’re recording this, and as is our habit, we want to come back and grade our own homework a little bit. We’ll check in on how those predictions are shaking out, where we got it right, where the world surprised us, and what we think the back half of the year holds in each of these areas.

Just a quick reminder on where we started. We said in January that if you were hoping things were gonna slow down this year, better luck next year. I don’t think anybody on this call is going to argue with that one. We don’t do this very often. We’ve got a really large collection of the team here with me today.

I’m very excited to have Russ and Joanna and Justin and Samantha and Ashley and Will and Lily all with us today. You’re gonna hear a lot of different voices, a lot of different perspectives on the topics that we want to discuss and kind of taking a look back and a look forward at the same time.

So we’re gonna go back through the big themes one by one, do a little recap on what we predicted at the beginning of the year, and get an updated read on what’s going on for the remainder of the year. So we’re gonna jump in and start with the two big elephants in the room that we led with last time, agentic commerce and stablecoins, before moving on to the rest of the topics.

So with agentic commerce, I’ll kind of set this one up since it was my prediction, back in January and my prediction was essentially that agentic payments, so an actual agent making the payment, wouldn’t be a meaningful channel for B2C merchants this year. I said the real action in 2026 is gonna be focused on product discovery, which would be merchants kind of figuring out how to make their products more accessible and more easily understood by chatbots so that they could even just get surfaced in front of consumers in the first place.

I leaned on the data that we’d had at the time, which was that chatbot usage was still, I mean, it’s growing quite a bit, obviously, like a super important technology. It’s evolving very rapidly. Back then, under 20% of chatbot users were using it weekly. During Cyber Week, in the fall in the prior year, we had seen about 20% of orders during Cyber Week that kind of touched an agent somewhere in the process.

And typically that’s gonna be really in that discovery component of the transaction, right? Where people are searching for products, evaluating different products, and then likely going directly to the site to make that purchase. And I said the heavy lift to get a merchant from zero to agentic payments because of the work that you need to do on that data side meant we probably wouldn’t see the payments element really showing up until later this year or possibly next year.

Now, Justin, I’m gonna throw this one to you. You’ve been looking into this space recently. How’s this prediction holding up at the midpoint of the year? Are we still in the discovery phase? Have we seen much progress on the payment side? Where are the developments looking on kind of B2C merchant agentic commerce?

Justin Pituch: Yeah. Great questions, Drew, and I think this is a super interesting topic. But we actually might be farther along than people expected in terms of the payment side of this in addition to that discovery piece that you were talking about as well, with some critical caveats that I’ll touch on as well.

So certainly true that product discovery is still an important focus for merchants today. Obviously, that opens up catalogs to chatbots being able to understand what a merchant has in their inventory. Bunch of different data points associated with that, things like star ratings, shipping time, et cetera.

But I think that there’s also a lot of meaningful movement on the payment infrastructure side. So we’re seeing some real progress in terms of protocol developments over the past few months that are really helping establish rules of the road for merchants, agents, and intermediaries. And that’s cool to see in this space.

We’re seeing a lot of work that’s interoperable, that’s open source. And so rather than having competing protocols and standards, we’re seeing real cooperation across players that eventually will help drive adoption here. So I think that what I’ll say here is that we’re seeing the conditions for adoption of the actual payment embedded in that agentic journey really coming into being.

But I think that the other side of that is that volume is a long ways off. So the capabilities are here, but mainstream usage isn’t. And I think that there are sort of two reasons for that, both on the consumer side and then on the merchant side of the equation. So consumers, and we’ve talked about this before, particularly younger consumers really like using chatbots as a search replacement, and they’re becoming more open to letting chatbots make purchases on their behalf.

There’s some data out there that speaks to that growing openness to that sort of transaction, but it is a pretty serious change in behavior. It’s a lot to ask a consumer to put that level of trust into the hands of a robot. So again, not really clear when consumers will be comfortable with that.

And then you have the problem of the merchant side, where merchants still face open questions around how do you attribute agentic orders properly? How do you manage disputes? How does loyalty come into play here? How do you make sure that you’re preserving customer relationships and making sure that customers have the experience that you want them to have even as they’re going through this third party?

And how do you decide… something that we’re thinking about a lot is, how do you decide what you’re willing to pay to generate those agentic sales? This is another place where I think the data is really inconsistent in terms of how well these sort of transactions convert and what the economics of these transactions look like.

So some unknowns there. What’s useful is that a lot of the standards work is being spearheaded by platforms like Shopify and like Google. So this isn’t necessarily a merchant-by-merchant effort. Each of those, if you think about Shopify, if you think about Google Shopping, bringing thousands of merchants into the fold every time they make progress in this space.

So I think, again, we might be further along than we anticipated we would be, but there’s still a lot of unknowns here. And again, I think the midpoint read is discovery is still really important here. It’s probably where merchants are paying most attention, and rightly so, but I think that there’s a good amount of progress being made outside of that in terms of the actual payments infrastructure.

So it’s certainly a space that we’re gonna continue watching really closely here, and continue watching both in terms of how are platforms starting to leverage this sort of technology and how close are consumers getting to actually being comfortable with it. Again, I think that’s probably the biggest variable here.

Drew Edmond: Yeah, and I would say there’s a still some major open questions related to the liability when it comes to these transactions that are also a bottleneck for merchants even letting bots in through the door to allow these payments to happen. So I think until some of those conversations are had and we have more clarity on that liability question, the trust component of agentic commerce is a challenge.

Russ, I think you had some comments.

Russ Jones: For me the most insightful thing I saw in the first half of the year about agentic commerce was a blog post from Dan Rosen, who’s the founder and partner at Commerce Ventures. His blog post was entitled “Agentic Commerce is a Zero Billion Dollar Market”.

Drew Edmond: Exactly.

Russ Jones: So protocol excitement aside, to your point, it’s very, very, very, very early in real payment volume, if any payment volume.

Justin Pituch: Yeah. We did some tests here. OpenAI has been thinking through this and thinking through how to get transactions happening in ChatGPT. And a few of us here at Glenbrook tried to buy something through ChatGPT a few months ago, and it was impossible, even though the capabilities were supposedly there.

So I think absolutely right. Probably not a lot of volume, but I do think that the protocols that are being established today, Drew, are addressing those key concerns that you brought up around trust and around permissioned access.

Russ Jones: I’m seeing all the time people are saying, “Well, what’s missing is trust.” I’m not so sure that’s really on the point here. I mean, let’s just say that trust is pervasive. Would this thing be exploding right now if trust was pervasive? I don’t think so.

It’s for exactly the points that you were talking about, Justin. Things aren’t quite ready yet.

Drew Edmond: I think it’s a necessary but not sufficient component.

Russ Jones: That’s exactly right. That’s exactly right. And maybe just as a public service announcement, should people be directly entering their card number into ChatGPT?

Is that good, Justin, or bad?

Justin Pituch: That is bad, and I think that’s the subject of Drew’s next question here.

Drew Edmond: Yeah, so staying on the agentic thread. Simon who wasn’t able to join us today, his prediction was that as the networks continue this push to remove card entry, manual card entry, and replace it with network tokens, that tokenization really becomes an enabler and a prerequisite for agentic commerce.

So Justin, just maybe you can touch on that point just to kind of close the loop here in terms of how do you think about Simon’s insight that agentic commerce both drives tokenization and, vice versa, that tokenization is a kind of a prerequisite here?

Justin Pituch: Yeah, I think that’s absolutely true. So Simon’s call definitely holding up. I don’t really have a splashy update for you though because of the fact that we’re not seeing a lot of volume here, if any, to Russ’s point.

So obviously tokenization is an important component of agentic because of the fact that robots should not be running around with your full card details and free to use them as they see fit. We want to equip the robots with credentials that are permissioned, auditable, and tightly controlled.

Then on the other side of the same coin or token, if you will, is that networks see tokenization as a vital security tool. And that explains, I think to a certain degree, why networks with their interest in expanding the reach of tokenization have been really keen to support agentic commerce from an infrastructure perspective.

As they see it, every agentic card transaction should naturally be a tokenized transaction.

Russ Jones: To start with, they think every agentic commerce transaction should be a card transaction.

Drew Edmond: That’s right.

Justin Pituch: Sure. Well, yeah.

Russ Jones: Let’s start there, and then let’s go to tokenization.

Drew Edmond: No, it’s a great point, Russ, because I think that whole conversation has kind of this implication baked in that cards are gonna be the only thing that we’re using in these transactions, and I think we know that’s not going to be the case, especially if we look at it from a global perspective.

There’s a lot of places where cards aren’t super popular to the extent they are in the US and other card-dominant economies, and consumers will want access to different types of payment methods. Maybe that’s buy now, pay later, installment loans, maybe stablecoins, maybe it’s faster payments.

Whatever it might be, it’s not going to be a card-only ecosystem.

Justin Pituch: Stablecoins and buy now, pay later though, those can be card transactions, and the networks want those to all be card transactions as well. So you could have a tokenized, agentic, stablecoin buy now, pay later purchase in 2027. That’s my new, net new prediction that I’m adding here.

Drew Edmond: There we go. There we go. But will it be an EMVCo token is the question?

Justin Pituch: Oh, geez.

Drew Edmond: The notion of a token may be used, but it’s not necessarily one that’s driven by the card networks, you know?

Will Eisler: Yeah, and I’ll just build on the agentic conversation, but from the commercial side. The consumer side of agentic is exciting and gets the headlines and, as Justin mentioned, has made some solid progress. But another area where it’s starting to grow is on the business side in finance back offices. And if you’d looked at this even a couple quarters ago, you wouldn’t have seen much, but it has progressed recently.

HPE’s CFO said in February that they’re scaling an internal finance agent across their transactional finance work, things like payables and receivables, after piloting it last year. So at least at some large companies, this has moved past the pilot stage. And Forrester’s view is that agentic payments will land in B2B before they land in consumer, because business payments are repeatable, and they sit inside guardrails that a consumer purchase just doesn’t have.

And Forrester expects about a third of B2B payment workflows to be using autonomous agents by the end of the year in things like invoice matching, reconciliation, and dispute handling. A reason it works there is really about the structure of the transaction. A consumer buying a pair of shoes is really a wide-open decision.

An invoice comes with a purchase order behind it, contract terms, and a tolerance band, so an agent can work inside those rules, clear an exception on its own, and leave an audit trail behind it instead of a chargeback. And the economics do push in that direction, too. Ardent Partners says processing one invoice by hand costs about $13 versus under $3 for the most automated teams.

The one piece that’s still early is the payment itself, as we discussed, but even that is starting to get built, as Justin talked about. So for the rest of the year, I think agentic headlines stay more on consumers and ChatGPT, while the more useful progress is happening in places like accounts payable.

And just something else I’d keep an eye on is what that does to commercial card and virtual card, because once an agent is choosing how to pay an invoice, it’s weighing things like working capital and rebate, and habit doesn’t really come into it. That’s a different kind of buyer than the commercial side has dealt with before.

The one caveat is that an agent is really only as good as the vendor data underneath it. The data is still pretty fragmented at most companies, and even the teams that are live, like HPE, built their agent to give the same answer every time with people still in the loop. So there is a gap between where this is heading and what a typical finance team can run today.

Drew Edmond: All right, so let’s jump now to the second elephant in the room. It’s been kind of 1A, 1B going back and forth, the fight between agentic commerce and stablecoins for the last 18 months or so. So now we’re on stablecoins. All right, Russ, back in January you said it wasn’t an elephant in the room, it was horses at a full gallop, maybe a stampede.

I think, 2025, we saw a lot of smoke, but we weren’t sure if we were seeing the fire yet. We had the GENIUS Act, of course, that came out putting regulatory stake in the ground, which really opened up the industry to kind of take more bets and for a lot of the work that had been kind of going on behind the scenes to emerge into the public view, with that regulatory change.

Explosive transaction growth tempered by the realization that actually very few of the transactions, when we look at the data, were actually payment transactions. We saw a wave of infrastructure announcements from a lot of different payment providers around the world, a bunch of companies that wanted their own stablecoin.

You have Western Union and Fiserv and Sony, and even the state of Wyoming that wanted their own stablecoin. So the big question that we had going into this year here in 2026 is that are we gonna see some real use cases emerge? Our initial read being that they’d be cross-border rather than domestic.

So before I get to you, Russ, I want to start with Ashley, who’s been deep in the digital currency weeds. Are we starting to see the fire or is it still mostly smoke? How are we seeing some of the use cases emerging? Is it what we thought they would be?

Ashley Lannquist: We are still waiting to see stablecoins being used for large-scale payments. We don’t have any data that is pointing to increased usage for real payments on a kind of global level. There are stories here and there of stablecoins being used in niche use cases, but what we were describing earlier in the year does seem to still be the case in terms of low usage for real payments outside the crypto ecosystem.

It’s still the case that almost all stablecoin transactions are for moving money between crypto wallets of large institutional holders or exchanges, or for trading within decentralized finance applications, for instance. However, there are still exciting developments in the stablecoin ecosystem.

We’ll see if these scale up to real usage, and as you said, so many companies make announcements, and that’s been happening in the cryptocurrency ecosystem for the past 10 years. So we have to separate this out from what really takes hold with end users. But there are some interesting stuff going on.

So personally, I’m quite interested in money transfer operators that are allowing easier cash out of users into local currency from stablecoins. So for instance, someone in the US could send a family member in a low-income country a remittance, and they could send it in stablecoin, and MoneyGram will convert it into local currency at the MoneyGram agent in that country.

So that end recipient doesn’t need to deal with stablecoins at all, which is important because stablecoins really do have a high barrier for usage that can be challenging in developing economies. You need to have the internet. You need to have a government-issued ID to have a virtual asset service provider account, which is a crypto exchange account. You need to have a pretty good sense of how to use stablecoins. So now these end users don’t need to use stablecoins to receive money.

Western Union has also been active in this space, as you mentioned, Drew. So they have a new stablecoin with Solana blockchain called USDPT, and it’s for their internal use. So I bring up this example to reflect another trend, which is not end users using stablecoins, but rather international money transfer operators using stablecoins to try to see, we’ll see really how strongly this adds value, but moving money within their own books and their own offices internationally with the stablecoin over the blockchain, in this case, Solana. So the end user could send from their local currency, and the recipient can cash out similarly in their local currency, and then in between, Western Union is using the stablecoin to get the money over to the agent in that country.

That’s quite interesting and we’ll see how that continues to develop. We do also see in the more enterprise side of things continued development. So Stripe with Tempo is very active in this space, and in March, it launched its private blockchain for high throughput stablecoin transactions, Tempo.

And this service is seeking to allow financial institutions, there’s a lot participating, to offer services with stablecoins at low fees and high throughput for customers. And it is also taking steps to make the experience better for end users. So for instance, it allows a financial institution that’s participating, like Nubank for instance, to sponsor the transaction fees of a stablecoin payment on behalf of the customers, the end users.

So every time a stablecoin transaction is made, generally, a cryptocurrency transaction fee has to be paid, and these new providers are abstracting away that for end users. So I think overall we’re seeing increasing ability to make this more usable for users and more reachable to developing economies.

Drew Edmond: Great. Thanks for that, Ashley. And Russ, given it was your prediction, how are you feeling in June?

Russ Jones: I think my prediction’s holding up pretty good. The real sort of, I guess, observation going into the year was there’s a ton of smoke here. There’s product announcements left and right. You can’t be a payment provider of any sort in the marketplace today without a stablecoin strategy.

But to Ashley’s point, we’re still waiting for the real breakout use case. At this stage it’s like a part of the feature set, we need to have it as part of what we do, we need to check the box for our investors, all that type of stuff, but we’re really still waiting for that breakout use case. And we still might find it.

The other thing that I watch very closely is the evolution of non-USD backed or non-USD pegged stablecoins. And there’s been a fair amount of things happening outside the US, particularly in Europe to start to propagate, I don’t know how you want to say it, Euro-backed stablecoins.

So one of the things that makes stablecoins so attractive to end users is there’s no FX. Everything’s pegged to the US dollar. It’s like, wouldn’t it be a simple world if every country in the world used US dollars? And yeah, it would be simpler, but that’s not the real world.

Drew Edmond: Absolutely. Staying relatively close to stablecoins, I think the emergence of stablecoins as an opportunity to move money around the world kind of creates an equal and opposite reaction in the ecosystem. And in January, Simon talked about the maybe lesser discussed cousin of stablecoins with tokenized bank deposits, which are blockchain-based dollars with a direct claim on existing funds.

He said the concept had been around a while but was gaining visibility with origins of private blockchains inside the walled garden of bank clients, and that we were starting to see banks enable them on public chains. And the open questions were whether usage would break out from institutional clients, large institutions, and whether tokenized deposits would end up as an alternative to stablecoins.

We also made the point that the economic leverage really sits on the tokenized deposit side, given the rules in the GENIUS Act that state that stablecoins have to be backed one-for-one with reserves, while bank deposits can continue to be lended against, as they typically are.

And so Ashley, given your view on the stablecoin lens, you’re the natural person to take this one as well. Where are we with tokenized deposits these days? Is it something that’s kind of growing within the banking system as a reaction to stablecoins and where do we see this kind of progressing?

Ashley Lannquist: I’d say that there’s a similar arc taking place here, although tokenized deposits are newer. So still very new. We still don’t see a lot of institutional or let alone retail usage for payments, but it’s earlier days. So there’s continued experimentation and prototyping and banks activity in this space. We’ll see how much they get used in terms of new activities in this space.

There’s a new article that came out in the Wall Street Journal just yesterday with another announcement with JPMorgan, Citibank, Wells Fargo, and Bank of America, where they’re going to develop a private blockchain for tokenized deposits transfers between those financial institutions.

The clearing house is involved with this. So this would allow kind of the benefits that you often hear about this digital currency, 24/7 transfers between these banks. It will launch, if all goes well, in the first half of 2027. So this could be promising. It’s another announcement with a lot of big players involved and we’ll see to what extent it’s used.

Another large announcement that has happened in the tokenized deposit space is not a live or even in development large scale platform, but rather a prototype, but it’s important for a couple of reasons. It’s called Project Agora, and it’s hosted by the Bank for International Settlements, and it has the IIF involved and a handful of central banks, including Swiss, French, European Central Bank, Japan, England, New York Fed, and Bank of Korea. It has a couple of big banks, BNY, BBVA. It has Visa involved and Swift as well.

And what this does is, it’s a prototype that looks at technical and legal aspects of having a shared DLT ledger with tokenized deposits from those multiple banks. And that DLT ledger connects with a couple private DLT ledgers operated by the central banks that put their own central bank reserves in a tokenized form on their central bank private ledger.

A central bank’s tokenized reserves is otherwise called a wholesale CBDC. So what this project does that’s new is it tests whether the banks in their actual ultimate institutional settlement of the tokenized deposits once the banks develop net debit and credit positions between each other, they can settle with central bank money, which is the tokenized reserves with those connected platforms.

So it shows the technical feasibility of those connections as well as legal feasibility. However, there’s still a lot more to explore here. And in all these cases of banks using tokenized deposits and central banks being involved, we do have still a couple thorny issues to address. A big one is data confidentiality.

So whenever you’re using private blockchains or public blockchains, there’s a challenge of preserving the privacy of the data. In a public blockchain, everyone can see the data. In a private one, you can have a little bit closer calibration of privacy and confidentiality to allow certain actors to see certain data.

But it’s not always that easy, and the default and starting point is that all of the validating nodes, which is the banks, which is the central banks in this case of Project Agora, for instance, can see the transaction information of the commercial banks. This can violate local laws and regulations and, of course, can not be in the interest of all of these financial institutions and their customers.

So we’ll really see how these thorny issues play out along with other ones like cybersecurity and data resilience. So to wrap up on the tokenized deposits front, still very new. I would say go easier on this sector because the banks are really trying to see what works here. There’s a lot of continued bank excitement and new technical work coming out, but we’re yet to see what develops into actual usage that’s valuable.

Drew Edmond: That’s great. Will, I’d love to hear you chime in on, I know you’ve been poking around and digging into the stablecoin space and the instant payments world as well, and kind of merging those two worlds together. I’d love to hear your thoughts.

Will Eisler: Thanks, Drew. I would just add a couple points. A lot of the real action here is in emerging markets, especially around cross-border payments and access to US dollars. In a lot of these countries, stablecoins have become a practical way to move money across a border and to hold dollar value when the local currency is sliding.

And just to share some data, Chainalysis noted that sub-Saharan Africa took in about $205 billion in on-chain value in the year ending last June, up about 52%, with Nigeria alone close to $92 billion. And that figure is all crypto, but stablecoins is a big share of it, and they’re what people use to move money across borders.And a big part of that is cost. Sending $200 into the region still costs close to 9% through the formal banking system, about the most expensive anywhere in the world. So a dollar stablecoin on a phone is just cheaper and faster.

The point I find interesting too is how this sits next to the topic of instant payment systems. A lot of these new domestic systems are local currency rails built to strengthen the local currency and bring more people into the formal economy. A US dollar stablecoin is pulling in a different direction towards holding and moving dollars, so the two can end up a bit in tension.

And the IMF put out a paper on this in December. Their worry is that when people hold dollars instead of the local currency, central banks lose some control over their own money and capital flows. And Nigeria’s own attempt at a government digital currency, the eNaira, never really caught on with about 98% of wallets inactive while people kept reaching for US dollar stablecoins instead.

So the thing I’d watch from here is just how regulators handle that tension. They can’t easily ban something people are using for good reasons, but also might not fully get behind a dollar rail that pulls against their local currency systems they’re trying to build. So they might try to bring this activity into regulated channels rather than block it, maybe through licensed local currency stablecoins or through bank and network partnerships.

Drew Edmond: Great. Thanks for that. So Joanna, I’m gonna go to you. We wanna talk a little bit about instant payments, and I think, maybe we’ll look closer at the cross-border part of this conversation. Back in January we were discussing the interconnection of instant payments as a trend to watch, and that’s kind of a topic that’s already come up here in different ways in this conversation around just the interconnection between systems, stablecoins talking to fiat, different countries interacting with each other, different banks talking to each other.

So I’m curious what your thoughts are in terms of, you paired that notion of the interconnection of instant payment systems with the G20 cross-border targets. You noted that a couple of months before our January call that they acknowledged that they were gonna probably miss the mark on their own kind of ambitious targets there.

What’s happening with kind of the Nexus implementation? Are we still waiting? How do we solve the cost problem? Anything notable that we can highlight here?

Joanna Wisniecka: Yeah, let me start with the easier ones. Really, the solving the cost problem is gonna be a long game.

So taking the easy one first, Nexus Global Payments. There have been some updates. There has been some progress since we last talked. Two things I’ll highlight. So one, in February, Indonesia became the sixth jurisdiction to join Nexus. They were actually, previously an observer. Now they’re actually a full participant.

So just the interest in Nexus and what it can offer is growing. Also in February, another announcement was made, which is really an indication of kind of moving more into the implementation stage, is that the technical operator of Nexus, kind of what we call the platform, right, the technology, the operator, was announced.

So there was a procurement process, an RFP process, and there was a joint venture between Malaysia’s PayNet IPS system, and Singapore Nets that will actually provide that technical operating role. So now there is a set of scheme rules, and there’s a platform, right? An operating platform. Currently that technical build is happening now that that platform’s in place. And the go live, currently, is targeted for 2027. So that’s certainly something to watch.

Maybe one other thing worth noting on Nexus is that the European Central Bank is evaluating a potential connection to Nexus. So that would be huge, right? Because we kind of think currently of Nexus as Asia-focused. If we get into the space of Europe potentially connecting or utilizing Nexus, that’s notable.

Assuming that these targets hold, kind of back to your question on cost, the impact on affordability, it’s gonna take a lot of time, right, for us to see what exactly, that contributes to.

A couple of things I wanna note, though, in this broader cross-border space. So one thing that’s kind of providing some signaling, one thing to note is that Swift has made an announcement earlier this year that they’re paying attention to and actually doing an MVP on a retail payments cross-border scheme. What’s really notable here is the retail focus and that they’re focused explicitly on improving transparency to end users on fees before end users commit to a payment, right? So there’s really a recognition on Swift now that retail payments really matter, and solving the cost problem is important, so they’re looking at how they can contribute to that.

One other thing to mention in the space that I think is interesting. In March, Brazil’s Pix, launched in Argentina. So they’ve actually launched some cross-border merchant payments capabilities in the last couple of years. But this one is notable because it is a little bit different. It is specific to merchant payments, but Pix users can now make purchases in Argentina using Pix scanning QR code using their own banking app, right?

So some of these past implementations have required you to create a separate account, potentially with a fintech, to be able to use that functionality in different countries. This is actually whatever app you use to use Pix, you can use now in Argentina to initiate a payment.

And there is the transparency of fees component here as well. So the customer is shown all of the fees that they will incur before they actually confirm the payment. And Brazil is now signaling that they’re thinking about this for potentially other markets as well. And back to Nexus, they’re also exploring Nexus.

So, kind of all linked in together. Lots of activity. Again, what any of this will mean for affordability to the end user, which is what the G20 roadmap is focused on, it’s TBD, right? This is gonna take some time, but lots of things to watch. I’d be curious, Ashley, cross-border payments obviously is a big, big topic, stablecoin, digital currency.

What are your thoughts specifically when it comes to affordability?

Ashley Lannquist: We can really think about the contribution of basically private blockchains or public ones to enable digital currencies to be transferred for cross-border payments from a couple perspectives here that can translate into lower end user costs. So really they’re enabling a platform where money can move on this shared programmable environment.

An end user or a bank in one country could send a payment to an end user or a bank in another country directly over the shared platform with an FX provider that’s connected to the platform in some manner, in some step in the process. So that’s key. The FX conversion still has to happen. It’s unlikely to be cheaper simply because you’re using a blockchain or something like this. That’s still going to be a major cost driver. But we think about things like stablecoins or other forms of tokenized money being used in between the sending and receiving financial institution or this money transfer operator example that I mentioned, and that might help reduce some costs.

Really the details will still ultimately matter though. Like, where is the actual institutional settlement happening? Is that still happening in traditional channels, which might be the case depending on how the platform is being set up. Where is the FX happening? Can the countries that allow participation in this platform agree to regulatory rules like AML, CFT?

So there’s opportunities here, but we have to wait and see exactly how these platforms will reduce costs for end users. More exciting times up ahead.

Drew Edmond: Speaking of costs to players in the ecosystem, maybe less so on the consumer side, but sometimes it can trickle down, but oftentimes not. When I think about the notion of interchange, one topic that we were talking about a lot back in January was the Visa and Mastercard settlement.

Chris walked us through the proposed settlement, the roots of which go back 20-something years, related to selective acceptance of different card types, surcharging and discounting flexibility, the reduction of around 10 basis points for a period of time for merchants.

So the big question that we had at the beginning of the year was whether any of it is actually going to stick this time, because we’ve seen this happen before. We know that these proposals get struck down. Merchant groups lobby and argue that it’s not enough change and that interchange is only really a piece of the cost of acceptance, and that’s a number, that total cost of acceptance continues to rise.

Samantha, I’d love to hear from you. Where do we stand on that settlement now that we’re halfway through the year? Is it sticking around? Are we heading back to the drawing board? And where do we see it going forward for the rest of the year?

Samantha Gordon: Okay, glad you asked because this one’s changing in real time. We actually got an update yesterday. But just to remind everyone since there’s a lot to this. This case is an oldie but goodie. The settlement unveiled in November 2025 is the third attempt to resolve this case, which has been running since 2005.

And there’s parts that get mixed up in here, so I just wanna pause to remind everyone listening that there’s two pieces to this. There was the damages settlement, which was $5.5 billion. Here’s money for what you got overcharged in the past, and that one is largely resolved. But we’re still talking about the injunctive relief side, which is the structural side of the settlement and the part that changes how the networks actually behave going forward, and that’s what’s still being fought over.

So here at the halfway point, the big news is, and Drew, I can actually update this in near real time because the ruling came down yesterday, so glad we’re talking today. Judge Cogan granted preliminary approval on June 9th, calling the amended settlement more extensive relief than the prior one that had been rejected in 2024, and finding it fair, reasonable, and adequate.

So after twenty-one years of litigation covering twelve million merchants, the settlement is moving forward. For some context on how we got here, there was a hearing back in April on the 27th, and Cogan signaled that he wanted to move quickly, and notably said that courts should avoid a meat cleaver, what he called a meat cleaver approach to dismantling the current structure. So that was the tell. Chris’s call from January holds.

But the so what on the ruling is that preliminary approval is not final approval, and merchants are opposing this. That opposition isn’t going away, it’s just moving to the next venue. There’s already been word that merchants plan to appeal to the Second Circuit Court of Appeals if Cogan grants final approval. So we’re not at the end of this story yet. A prediction for the back half of the year is that we’re still following this for another few months.

So the merchant opposition piece is interesting. It was broad and organized going into yesterday’s ruling, and it’s staying that way coming out of it. They were obviously prepared for next steps if the ruling went the way it did. Walmart filed a formal objection asking the judge to divide the merchant class, basically saying that large national retailers’ interests diverged too widely from smaller merchants, and there were similar objections coming from the National Retail Federation, National Association of Convenience Stores, Restaurant Law Center, and some others.

So yesterday’s ruling didn’t make them happy. This is just moving their fight to appeals court. Their argument is that this settlement fixes one line item interchange in a cost stack that just keeps growing on every line. So to put some numbers behind that, the headline of this case has been the ten basis points of interchange relief, which sounds like something meaningful, and I don’t want to discount that. It is.

But the National Restaurant Association’s filing pointed out that when average interchange rates rose from two point two six percent to two point three five percent in 2024, there’s your ten bips right there. So cutting rates by ten basis points isn’t really relief so much as treading water. Another point is the one point two five percent ceiling on standard consumer cards for eight years, which actually is pretty meaningful.

But this doesn’t apply to premium cards and commercial cards, which is where the costs have been climbing fastest. And this case doesn’t touch scheme fees, assessment fees, auth fees, digital commerce, tokenization fees, et cetera, et cetera, which have also been increasing. This is real money, and it’s not addressed by the settlement.

So looking into my crystal ball, this settlement is sticking for now. For the merchants who are hoping that the court would force more structural change, yesterday’s ruling means that the legislative and the regulatory path is now the main game, and it’s gonna move there.

You know, if you look around the world, the EU caps interchange and scheme fees. Congress took a swing at that with the Credit Card Competition Act recently and missed. But if this settlement moves towards final approval while total acceptance costs keep climbing, this conversation is gonna get louder, and that might be moving to the next battleground.

Drew Edmond: So I think the equal and opposite reaction of rising total cost for merchants is surcharging, right? And their reaction is, “Hey, if these costs are going up, I need a way to mitigate these costs” in some cases. And the idea of who pays for payments, I think, is a constant global question that gets treated differently in different jurisdictions, right?

It looks different in certain African countries that we’ve been doing work in, different in the US, it’s different in Australia. Even within the US, right, we’ve got federal laws, we have different state laws, kind of this whole patchwork and minefield that everybody has to navigate.

And the compliance related to it has kind of become its own cottage industry in terms of businesses that have cropped up to help merchants both enable this and remain compliant with this challenging environment.

Maybe Russ, I’ll throw this to you. where are we right now in terms of, of surcharging? Last time we talked, when we were discussing surcharges around how consumers don’t love them, you said consumers hate surcharges, which I think is true.

Russ Jones: I stand by that.

Drew Edmond: Yeah, I stand by that too. Who likes pay that?

Russ Jones: Right on the money on that prediction.

Drew Edmond: Yeah, the Oracle. You’re not the professor, you’re the Oracle.

Russ Jones: The framework for surcharging in the United States, which is one of four markets in the world where surcharging is allowed by card networks through regulatory pressure. The surcharging framework in the US goes back, yeah, 10 plus years, which gave merchants the ability to surcharge credit cards but not debit cards.

And the proposed settlement that Samantha was just talking about has a big sort of surcharging overhang to it in that it creates more product, if it’s accepted and codified by the judge. It creates more product categories that can be treated separately from an acceptance point of view and surcharged separately from. So there could be some change on the horizon in the US.

But still in general, consumers hate it. Month over month over month, you see increasing uses of surcharges. The thing I would note in the US that I see changing is you’re seeing it starting to be incorporated as a feature inside of a lot of retail systems so that instead of the small merchant sort of willy-nilly surcharging and not really understanding what the surcharging rules are, you’re starting to see support for surcharging be formalized inside of a lot of software packages, which they’re formalized in a way that is compliant with card network surcharging rules. So that kinda says that we’re gonna see more of surcharging in the US, I think.

To your point about payments being different country to country though, Drew, the winds are blowing in the opposite direction in Australia, which has for 20 plus years allowed surcharging under government regulation, surcharging on credit cards, debit cards, prepaid cards, basically any sort of open loop card.

And that has created a context where it’s evolved to the point where every purchase done with a card in Australia has a surcharge added to it. And it’s like a nuisance tax. The interchange rates aren’t really gigantic in Australia, not like in the US. And still there’s a surcharge added to almost every transaction.

So the the Reserve Bank of Australia has set down new rules that are going to adjust interchange rate caps, and completely ban surcharging across all merchant categories, effective this October. So that’s exactly the opposite of what we’re seeing in the US market.

Samantha Gordon: Yeah. I think an interesting thing about the RBA’s approach is that the logic is surcharges no longer steer consumers to cheaper payment methods.

Russ Jones: It’s like, adding a charge for electricity on the bottom of the receipt.

Samantha Gordon: They’re saying this is just how people pay. This whole mechanism broke down as soon as credit card acceptance became universal. This isn’t doing what it was intended to do anymore. And I think, to your point, Russ, I think the US is becoming more and more alone on the island in the way our regulation is treating it as opposed to the rest of the world.

Russ Jones: Yeah. The other subtlety in how what’s happening this fall in Australia is they’re adjusting the interchange rate caps, which have been in place for a very long time. And they were based on weighted averages, and they’re shifting over to a model now that’s more– that’s absolute. And the weighted average approach gave payment providers the flexibility to price card acceptance to be more expensive for small merchants and more cost affordable, cost-effective for large merchants.

The Reserve Bank doesn’t view that as fair, and they’re moving to an absolute rate cap that’ll be, doesn’t matter if you’re the largest merchant or the smallest merchant in Australia, you accept a credit card, you’ll have the same rate cap

Samantha Gordon: Yeah. And I think taking it back to the US for a second too, because one of the things Chris pointed out six months ago is that the painful part in the US isn’t just consumers dislike or hate surcharges, it’s impossible to navigate here. I think Chris mentioned that we have a patchwork of federal, state, card brand, network regulations to navigate, this patchwork is a minefield, and none of that’s changing either.

Drew Edmond: Moving over to risk and fraud and VAMP, which has been a hot topic this year. Maybe, Samantha, we can just do a quick one here, just touching on, when Chris was talking about this back in January, he was mentioning that we were seeing global card fraud tick down a few basis points.

So I want to get your input on, is that a trend that we’re continuing to see? Do we have the data on that to, to give us an indication? Obviously, overall numbers are going up, but from a basis point standpoint, it seems to be slightly, slightly decreasing based on the data that was used during that conversation.

I think the VAMP conversation in January was really around kind of agentic commerce, and as we know, the volume on that is essentially zero. So my guess is that VAMP and agentic is not really a current concern. Could be in the future as agentic volume starts to take off.

But give us some insight on kind of where we’re at with overall global card fraud rates and maybe just VAMP more generally.

Samantha Gordon: Yeah, so I’ll give the good news first here because it has held up. We saw in the Nielson report, which confirmed in January that global fraud losses did dip, I think it was 1.2% in 2024. They actually credited AI tools with helping the industry build the best fraud-fighting models it’s ever had.

So the trend Chris was flagging is intact. Card fraud as a share of volume is moving in the right direction, and we hear a lot about AI fraud, but I think at Glenbrook, we’ve been beating this drum for a while that AI is a tool for the white hats or the black hats, so it’s nice to see AI tools on the good side too.

Yeah, VAMP is where everyone’s attention is. It’s gotten more potent, not less since January. VAMP is volume-based. That’s the change that keeps catching merchants off guard. So every dispute and every fraud report counts equally, regardless of transaction size, which means high volume, high ticket count, low value merchants are particularly exposed.

Thresholds tightened again in April, so the bar keeps moving down, and the consequences of landing on the wrong side are very real. It’s fines, holdbacks, potential removal from the network. The trap that caught merchants early, which was clearing disputes before they became chargebacks to stay under the whole thresholds, so that playbook is gone now. VAMP combines fraud reports and disputes into a single ratio, so the math changed on everyone, and that was a big adjustment to a lot of fraud strategies.

So agentic, though, is where this all gets interesting. As you said, volumes are still low. It’s starting to show up, but it’s still early. It’s complicated. Dispute data is showing a rising trend of customers disputing charges from agents, automated decisions they didn’t fully understand or expect, and this is a genuinely new dynamic, We’re still figuring this out, and in every protocol being proposed, the merchant remains the merchant of record.

So this means that the merchant is still taking on the risk of autonomous execution. But what we also see now is there’s a false positive problem running in the other direction, which we should not be surprised about, if we think about this kind of logically, that fraud systems have been designed for years around earlier technologies to block bots from making purchases.

And I don’t think it’s a huge shock now that in the early days of agentic commerce that legitimate agents are being misclassified as malicious bots and good transactions are getting declined and creating their own friction and dispute chain. But this means that merchants are getting squeezed from both sides. Agent-driven purchases are generating disputes that they can’t defend that easily, and their own fraud tools are blocking legitimate agent traffic.

Drew Edmond: Well, for better or for worse, with agentic commerce being so limited right now, I think, hopefully it’s not contributing too much for merchant pain in the world of VAMP.

Samantha Gordon: And if people are looking for more information on VAMP, we did a whole episode on this back in April. It was episode 290.

Drew Edmond: Let’s shift over to the regulatory climate here at home, more focused in the banking ecosystem. Back in January, Bryan was talking about the fact that in prior years, 2025, 2024, we were talking a lot about regulation. We were talking a lot about issues happening in the banking ecosystem with like fintech banks that were causing a lot of issues.

And the specific thing that Bryan was watching this year was the return of de novo banking. And new bank chartering had barely really happened in kind of last five to 10 years. And then all of a sudden in 2025, it started picking up steam. We had fintechs and non-banks going for limited purpose charters, new national trust charters, some fintechs going for full national banking charters.

So his thesis was that this could push the US to look more like the rest of the world. You look over to Europe with e-money licenses and other places with lighter weight charters, which could lessen our dependence on the somewhat uneasy sponsor banking model that had emerged in the US.

So Lily, you’re taking this one, so I’ll hand it to you. Is the de novo trend playing out the way that Bryan described, or does it still feel like the year of the regulator is behind us? Where do you see this world of chartering and the broader regulatory climate heading in the back half of the year?

Lily Abese-Gihozo: Thanks, Drew. Brian called it right directionally, but the pace has actually exceeded what most people anticipated at the start of the year. The OCC has already approved four new applications and received more than seven just this year. In combination with the fourteen applications they received in 2025, the total nearly matches the total application received from the prior four years.

For the full national bank charters, Nubank received their conditional approval in January and Revolut applied in March. For ILC charters, we’ve seen conditional approval for Ford Credit Bank and GM Financial in January, and Stellantis got their approval in May. For national trust charters, this is where majority of the activity is in. We’ve seen applications, close to eleven thus far, with Circle and Ripple currently having conditional approvals as well.

I would agree with Brian that having more chartered banks lessens the dependency of sponsored banks in the long term, but I do believe there remains a significant opportunity for sponsored banks, particularly when serving earlier stage startups that prioritize rapid scaling and innovation.

For these companies, obtaining a bank charter can introduce substantial regulatory and compliance obligations that may divert resources and attention away from their core growth objectives. As a result, the sponsored bank model continues to provide meaningful value by enabling fintechs to access banking infrastructures without having to take the full burden of operating as a chartered institution. This dynamic is likely to sustain the demand for sponsored banks, especially for emerging fintechs. And we’ve also seen with fintechs pursuing bank charters earlier this year, the majority were already operating at a considerable scale before taking this step.

In regards to is this the end of the regulator, I would say that it’s not the end of the regulator, but rather a shift towards a regulatory climate that is friendlier to entry, but the standards for actually opening remain high. The OCC conducts a deep assessment of the business plans, the operations, and risks before issuing conditional approval. I wouldn’t say we’re quite yet becoming Europe, where e-money licenses create a more standardized middle tier between the bank and the non-bank. I would say that the US is doing something more fragmented with the menu of charters, the conditional approvals, and the state and federal routes that they can take.

In the later half of this year, I’m personally just watching the conditionally approved banks to see if they get their full charter and begin operating and how they shift the market dynamics as we see them today. The GENIUS Act has a July eighteenth rulemaking deadline, which I infer will drive another wave of trust charter applications from stablecoin issuers who now have more regulatory clarity.

If I could update the call, I would say the US banking regulation is entering an era of regulated access rather than deregulation. This theme is further reinforced by the proposed PACE Act, which came out in April, which would provide entry for payment companies into the Fed services, and Executive Order 14405, which is called Integrating Financial Technology Innovation into Regulatory Frameworks back in May, which aims to review the current policies, rules, guidance, and application processes and to identify if there’s any unnecessarily blocking fintech partnerships, charters, or licensees and payments access.

Drew Edmond: Okay, so it’s not the Wild Wild West completely, but we’re making it a little bit easier for folks to take on some of these new charters and that will ideally lead to some interesting innovation. Great. Thanks for that insight.

All right. So well, thank you, everybody. Lot of topics to cover, and more that we could have covered, but we ran out of time, so we’ll have to save this for another one in the future. Thank you all for joining me today. Thank you for the conversation, and thank you to our listeners for listening. Have a great one.