# Stablecoins as the Next Financial Infrastructure Layer

**Podcast:** web3 with a16z crypto
**Published:** 2026-02-02

## Transcript

I don't think it takes like that much of a stretch of the imagination to believe that in five years or 10 years, the overwhelming majority of the payments that happen in the world just by sheer number of payments are happening via stable coins and predominantly happening between non-human agents.
What are all the things that need to be built when the velocity of money goes from X to 10x or 100x?
What's all the subsequent infrastructure that needs to be built around to support those use cases?
It feels like it's coming.
Zach, thank you for being here.
Excited to be here.
What is your pitch to people who are not in crypto who maybe don't get crypto, don't understand it, don't care about it so much.
How do you describe what you're doing and the value that you provide?
I think of stable coins as an evolution of financial services.
You know, the last evolution of FinTech was a lot of companies building on top of cards.
Cards were this new form factor for money that enabled money to move much differently than it could before.
Cards sort of changed the way that, you know, commerce happens, and a lot of the companies that were successful in FinTech were building on the back of cards, the stripes and squares and audience and so on.
And stable coins are sort of the next evolution of money.
They're sort of just like cars, they're a new form factor of money.
And just like cards enable a whole new set of capabilities with money, you know, that can be moved cross-border very cheaply.
It can be programmed, it's natively yield-bearing, it can be moved in very small increments.
And just like cards, it will evolve to enable a whole bunch of new money applications.
And we're starting to see that today.
A lot of people forget this, but in the early days of credit cards, it reminds me of the early days of the crypto industry.
You know, there was a lot of bad, bad, bad headlines.
You know, people were airdropping these things all over the place to people who had terrible, you know, they didn't know they had terrible credit, but they found out that they did, and there was all of this fraud, and um uh it was a it was a really tough go.
And actually there was a whole moral panic around them too.
Like senators were saying this is gonna cause people to get into debt they can't get out of.
And now today you fast forward decades later and the whole system's based on cards.
Exactly.
I mean, and there was the same adverse selection dynamic of, you know, okay, who's gonna use cards?
Only the people who can't pay for things.
And then there was a similar view of like who is using, you know, stable coins and crypto assets, oh, only people who can't access other financial services and financial rails.
And in both of those cases, the card effectively, you know, grew over a 10, 15 year period and scaled sequentially into more and more and more use cases, but really accelerated when this other sort of big change happened in the world, which was you know, the internet.
And we all use cards before the internet, but you know, it wasn't the same as it is now.
And I think the same thing will be true with with stable coins.
You know, stable coins enable all these applications today, but we are like just scratching the surface of what is possible.
And with AI and a agentic payments emerging, I think that the same thing will happen where this platform, this sort of one technology platform will accelerate this financial services platform and vice versa.
So there was a time uh during this adoption phase, people were afraid to to enter their credit card numbers on the internet.
They thought that it would, you know, you'd get all your data stolen, uh, people would drain your bank account or something.
When it comes to stable coins and where we're at now, what are the challenges toward, you know, obviously they're being adopted quite strongly across the board, but are there any obstacles or hurdles to overcome to really fully make it the next platform that everybody uses?
I mean, it's hard to overstate just how early we we still are.
Bridge started three and a half years ago.
We launched our APIs two and a half years ago.
We were, you know, one of, if not the first company to build APIs that enable the companies to build stablecoin applications.
So then definitionally that means that there's sort of only been, you know, companies building a lot of these stablecoin applications for two and a half, three years.
We're like just at the very beginning of figuring out, you know, all the applications that can and should be built on top of this stablecoin platform.
And as a result, you know, we're just the beginning of figuring out what infrastructure needs to be built to support all this.
An example for that for us is like our product itself has evolved very wildly over the last two years as we've moved from you know early developers who are using the product, like you know, our first customer was Zulu, which was moving money cross-border, and everything was like kind of manual, and we were sending like two or three payments a day, to then shifting to the US government that was sending aid payments and sending thousands of payments sequentially, which required a whole new set of infrastructure to be built, and then to banks and like large-scale financial institutions, which then require a whole new set of regulatory infrastructure and messaging systems to confirm payments and rebalancing of treasuries.
And so we're just beginning to figure out how people want to use this new infrastructure and how or how agents will use this new infrastructure.
And as a result, what needs to be built to facilitate the use cases that people have in mind.
What use case or application are you most excited about?
I'm perpetually most excited about the next one.
Uh it's hard for me.
Like looking back, the first couple months of the company were kind of defined by the cross-border payments use case.
And that was like the one that I was most excited about.
And then this like aid disbursement and mass payouts use case came, and and that was like the US government, but then also scale.ai.
Scale was paying out people who were doing data labeling all over the world with stable coins.
And then the neo-banking opportunity came along, and folks started building neo banks on top of stable coins, serving a bunch of countries that were not served before.
And then SpaceX came along, and this was the treasury use case of moving money cross-border with stable coins.
And then now we're seeing, you know, card and payment use cases, and uh, and we're, you know, our stablecoin issuance business is starting to scale.
I would say every probably six months or so, like the the use case that I'm most excited about is a new one that we weren't even thinking about.
Stripe is enabling people to use stable coins at checkout, and the interchange fee is a lot less than you would get on a traditional card network.
It's like half 1.5%, I think is what you guys are charging.
So we're already seeing this in practice and also seeing like other venues like Shopify apply the same thing.
I don't think it is it takes like that much of a stretch of the imagination to believe that in five years or 10 years, the overwhelming majority of the payments that happen in the world, just by sheer number of payments, maybe not by volume of payments, are happening via stable coins and predominantly happening between you know non-human agents, and then like what what are all the things that need to be built when the velocity of money goes from X to 10x or 100x?
And what's all the subsequent infrastructure that needs to be built around to support those use cases?
We're not there yet.
We're beginning to just see the green shoots of these types of opportunities, but it it feels like it's it's coming.
You mentioned agents.
Do you see this as the next step or one that is further out?
The great benefit that we have is that we don't necessarily have to be opinionated around what use cases are going to work or what use cases are not going to work.
We kind of have built this infrastructure, and then developers come to us, and then they show us the future.
And so we end up having a glimpse into the future, maybe six months or a year, year and a half ahead of others as we start to see, you know, all these different founding teams and enterprises coming to us to build stuff.
And right now we are just starting to see a surge in demand for agentic type applications.
We're seeing a bunch of new companies that are formed that are building on top of our APIs to build different agentic payments use cases.
We're seeing folks come to us to issue stable coins related to agentic payment use cases.
And then we begin this like creative process where all these different folks try with their own variations to solve this problem, and then a few end up winning and then they end up scaling.
Is there a particular area that you're seeing likeliest to be the early adopters?
Right now, a lot of what is being built with agencommerce is uh buying things in a chat interface predominantly with a card.
And I think that's sort of like, you know, this is like a human searching goods and acquiring a thing.
And that probably will not move to stable coins.
Maybe it will, maybe it won't, but but I'm less optimistic about that.
I'm much more optimistic about you know what Cloudflare is doing, which will enable true sort of like agent to agent without a human involved at all.
Micropayments for access to information.
And I think this is like extremely important for the internet.
One of the things I'm I worry about a ton with you know the emergence of all these chat interfaces is that basically are sucking all the information out of the internet uh and pulling it into one interface, but it but it makes it hard for other people to want to create additional information on the internet.
And you need to have some new economic model when the web pages themselves are disintermediated from the consumers of those web pages.
One thing that the card networks argue is that you get, you know, all this ability to combat fraud, they can deal with chargebacks, things like that.
Um, and that stable coins aren't ready to tackle that.
Uh and that maybe that's priced into the interchange that they that they charge.
Will stable coins get there?
Are these features that you imagine will be added to stable coins they're missing now?
Will they become more expensive then?
Yeah, I I think that there will be an increasing ability for wallet-to-wallet payments.
This is what Stripe enables today.
So you have your phantom wallet, you show up to Versel or something, and you want to pay with stable coins, you could directly pay.
I think that will grow and that will be very low cost, you know, tens of basis points, maybe even less to facilitate those payments.
But I also don't think those will be wildly popular.
There is a reason why cards are as popular as they are, and that level of you know, chargeback and settlement and fraud prevention is very important.
However, beneath the card, the overwhelming majority of the fees come from banks, not from you know, visa and like those benefits that you really want.
So, what I think will happen is that there will be the emergence of infrastructure that provides just what you're talking about, you know, the fraud and chargeback prevention and so on, but gets rid of all the bank fees because you no longer need a bank when you have a wallet.
And as a result, you will have a payment method that has all the benefits of a card, but is one-fifth to one-tenth the cost and enables instant settlement for the merchant.
So it should be a better payment method all around.
Stripe also has been now, you know, they were sort of early to applying crypto to their business, and they've gone through various cycles of trying experiments and sort of ditching them.
Originally they allowed you to the make purchases with Bitcoin, and then there was the whole stellar phase, that was a thing.
But now it seems like they're leaning really strongly in.
They're assembling various pieces and parts, you know, acquiring bridge, acquiring privy, working on tempo, their own blockchain infrastructure.
What is the end state of all this?
What is Stripe's ambition here, assembling all these parts together?
And what does it gain from owning all those various parts of the stack?
It's hard to say, you know, where all this leads because we're at the very beginning, like I mentioned before, we are just so phenomenally early in the evolution of this space.
But what has become incredibly clear to us is that there are core bits of infrastructure that are going to be disproportionately important for all the folks who are building the applications that will be successful with stable coins.
And like Stripe has sort of built its entire business on, you know, being a developer-friendly means to accept card payments, you know, the accept card payments and five lines of code or what what have you.
And the same thing in the same levels of abstraction need to be built for stable coins.
And in order to build with stable coins and to build payment applications on top of blockchains, you need sort of bridges infrastructure, which facilitates the money movement.
You need privy's infrastructure, which facilitates the store value.
And in some cases, you will need the tempo infrastructure to facilitate the velocity of money and the use cases that are needed.
And so in each case, we're sort of investing at the layers of infrastructure that enable builders of any size, shape, to create the applications that we think or they think more importantly could be really successful.
Stripe has talked about, and you've talked about needing to keep things open, uh, especially when it comes to tempo, that nobody will build on something that they think is gonna like close.
But if you look at history, there are other networks in finance that started open and became more closed and consolidated over time.
I think visas like the classic example people would point to.
How do you ensure to people that this won't follow the same path?
So, first off, just like at a high level, strongly agree.
I mean, we believe that we're creating tempo because we think that a payment scale blockchain is important and a missing bit of infrastructure.
We also believe in order for Tempo to be successful, uh, many folks who are building payments applications need to want to build on top of it.
And the only way that folks will want to build on top of it is if they believe that this infrastructure will be best for their business today and long term.
And then the only way they believe will be best for their business if they believe it will be neutral and will be facilitating making decisions that are in the best interest of not one party, but the totality of folks who are building on top of tempo.
So very, very strongly agree with that.
And as a result, Tempo needs to be decentralized.
It can't be, you know, dictated and the roadmap dictated just by Stripe or just by any one entity.
And that is a path that we are moving down very aggressively.
Yeah.
Chris Dixon likes to talk about this sort of difference between phases of the internet and how like the previous era was don't be evil, Google's classic tagline, and now with blockchains, you get these computers that can make commitments, and so you get assurances that things can't can't be evil, can't become evil if they're sufficiently decentralized.
Yep.
I mean, this was like a big a big decision in like the whole, you know, L1, L2 conversation around tempo is that it is currently challenging to decentralize L2s, and it is there's a very clear path on how to decentralize L once.
And so that was like a core tenet of the of very early decisions that the team made as they were investing in tempo.
Well, we definitely want to have you come back and tell us how it is all coming together.
You have time for a lightning round of questions?
Yeah.
Okay.
What's the worst advice you've ever gotten as a founder?
Many things.
So the the first is hire really senior people.
Second is you'll know when you have product market fit.
You didn't know when you got it?
No.
I I honestly I still question whether we have product market fit.
Oh wow.
Yeah.
I think that this whole narrative around there's pre-product market fit and post-product market fit companies, and you could clearly draw a line down the middle and separate the two into two clear categories is completely wrong.
And I think it's like a total disservice to founders.
Wow.
Yeah.
All right.
What book should people read?
As many as possible.
I genuinely like I like sci-fi.
There is a ton of sci-fi that I read, and there's a lot that I learned from it.
There's obviously business, but like I'm reading John Malone's book right now, and there's a lot that I take away from the evolution of the cable industry that is very similar to the evolution of our industry, like just general fiction books.
I've read a ton on like Chinese history, which was very, very interesting in a period of time.
And you never know what you're going to take away from a book.
And oftentimes the books that you think are going to be the most helpful end up being the least helpful.
What's the best book on Chinese history?
I wouldn't say it's the best book, but the book I just finished was Mao's Great Famine.
Yeah.
I just had very little appreciation for the, you know, all that sort of went on and understanding sort of the political machinations of China gives me a much deeper appreciation for for where the country is now and the decisions that they're making relative to the decisions that we're making.
Totally agree.
And on the sci-fi point, um, you know, the three-body problem series of just to combine with uh, you know, Chinese history.
If you're talking about the Great Famine, like there's a whole plot point in that book about mass starvation that happens coming from a Western perspective.
It's like, wow, that's really far out.
But then you think about where it's like the book is coming from culturally, and it's like this is something that happened in recent memory for people.
It's like really not that long ago.
It's seared into like a lot of the politicians like Xi grew up during this during this period of time, and then the subsequent the Cultural Revolution and uh and so on.
What's your biggest productivity hack?
Working out every morning.
Okay.
Every every day working out is probably the single most important thing.
I would say probably the single thing that I do differently than anyone else is every day I have two blocks in my calendar of two hours of continuous working time.
Oh wow.
With no meetings.
Undistracted.
Undistracted.
And finally, what is the smallest hill that you will die on?
This is like a bigger hill, I guess.
I would say, well, I would say a bigger hill that I would die on is that I don't believe in the role of management.
Okay.
Coming from somebody who's tasked with managing a company.
We we don't have any managers.
Uh we're 150 people without any formal managers.
How do you prevent anarchy and and chaos?
I think you can split management into sort of mentorship around a technical skill, and then all of the other things related to leveling and performance and so on.
And I think all these other things, if your company is doing well and everyone is really busy, then the first is what really matters, and the second becomes a side quest.
And by formalizing managers, you cement the side quest as a core part of your journey.
So just a quick question on that, a quick follow-up.
Is this a luxury of being earlier in your journey?
Like is this something that you think will have to change over time, or do you expect this is the way it's gonna be?
There is a long history of founders who have had the same view and who now have many levels of managers.
So uh am I the one founder, you know, that will figure out how to scale this?
I I don't know, but we are going to take it as far as we can.
This time is different.
Yes, yes, this time is different.
All right, Zach, thank you so much for your time.
Yeah, thanks for having me.