The Future of Finance Will Be Built on Infrastructure, Not Apps

Wispr Flow turns your voice into clean, professional text you can send the moment you stop talking. Not rough transcription you have to clean up. Actual polished text — ready for email, Slack, or any app.

Speak the way you think. Go on tangents. Change your mind mid-sentence. Flow strips the filler, fixes the grammar, and gives you text that reads like you spent five minutes writing it.

89% of messages sent with zero edits. Millions of professionals use Flow daily, including teams at OpenAI, Vercel, and Clay. Works on Mac, Windows, and iPhone.

Start flowing free

Last week, I wrote that the future of finance needs rules.

Regulatory clarity matters because institutions need confidence before they commit meaningful capital, build new products or change core infrastructure.

However, rules are only the beginning. The more interesting question is what gets built once those rules become clearer.

For most of the last two decades, financial innovation has happened at the application layer.

I think we're entering a different era. The next wave of financial innovation won't just improve how we interact with the financial system.

It will change the infrastructure the financial system runs on. It will create a much bigger opportunity. 

Think about many of the biggest fintech successes of the last 15 years.

They dramatically improved the customer experience.

Those were meaningful innovations, but underneath most of them sat much of the same financial infrastructure.

The application layer changed much faster than the infrastructure underneath it.

That's beginning to change.

Stablecoins are one of the clearest examples. It’s easy to dismiss them as digital dollars used by crypto traders. I think that misses the bigger story.

Stablecoins make money internet-native and programmable.

They can potentially move between institutions, applications and markets without relying on the same collection of intermediaries required by traditional payment systems.

That opens possibilities far beyond crypto trading.

Eventually this will include transactions initiated by AI agents.

The important innovation isn't that we created another way to represent a dollar.

It's that money itself can increasingly become part of software.

Money isn't the only thing changing. Assets are becoming programmable too.

Potentially almost any financial asset can be represented digitally and connected to programmable infrastructure.

We often describe this simply as tokenization, but tokenization isn't particularly interesting if all we're doing is creating a digital representation of an existing asset.

The bigger opportunity comes from what happens next.

Financial products that historically lived inside separate databases and institutions can potentially begin interacting with each other through shared infrastructure.

That's a much more profound shift than putting a stock certificate on a blockchain.

Today's financial system contains enormous amounts of friction because transactions often move through multiple institutions, databases and reconciliation processes.

A transaction might appear instantaneous to the customer while the actual movement of money and assets behind the scenes takes considerably longer.

That creates complexity, it creates counterparty risk, and it ties up capital.

Ultimately, it creates an enormous reconciliation industry simply to make sure everyone's records agree.

Blockchain infrastructure introduces a different model. The transaction and settlement can increasingly become part of the same process.

Markets can operate continuously. Assets and money can potentially settle against each other programmatically.Rules can be embedded directly into transactions.

That doesn't mean today's financial infrastructure disappears. Banks, custodians, exchanges and regulated financial institutions will remain enormously important.

However, their infrastructure can evolve.

Settlement increasingly becomes something software can coordinate.

This is where Ethereum becomes particularly interesting to me. Not simply as a cryptocurrency, but as infrastructure.

Ethereum provides a programmable settlement layer where money, assets and applications can interact through a common protocol.

Today, much of that activity still happens inside crypto markets, but the boundary between crypto markets and traditional financial markets is becoming less obvious.

Stablecoins already represent hundreds of billions of dollars of value.

Major financial institutions are experimenting with tokenized funds, deposits, bonds and other financial assets.

The question isn't whether every financial transaction eventually happens on Ethereum. It won't, but there will be multiple networks, private infrastructure and existing financial rails.

The more important question is whether Ethereum and other programmable networks become meaningful pieces of the infrastructure connecting global finance.

I think that's increasingly plausible.

Whenever infrastructure changes, new businesses become possible. 

Each shift produced companies and business models that were difficult to imagine beforehand.

Programmable financial infrastructure could do something similar for value.

Imagine financial products that can automatically respond to market conditions.

Businesses that can transact globally without assembling separate payment infrastructure in every country. 

Importantly, this will eventually result in AI agents capable of conducting financial transactions on behalf of people and organizations.

Many of these ideas still sound early, but that's exactly why infrastructure matters.

New applications become possible when the underlying infrastructure makes them possible.

There's another reason I think this transition matters.

AI is rapidly becoming capable of doing more than generating information. AI systems are beginning to take action.

Eventually, AI agents will increasingly participate in economic activity, but an AI agent can't easily operate inside a financial system designed around humans logging into bank accounts, manually approving transactions and moving information between closed databases.

Agents will need financial infrastructure they can interact with programmatically.

This is where the convergence of AI and programmable financial infrastructure becomes especially interesting.

We are still early, but I suspect this convergence will ultimately matter much more than another generation of financial apps.

As an investor, entrepreneur and capital allocator, I've learned to pay attention when infrastructure starts changing.

Once infrastructure becomes widely adopted, thousands of companies can be built on top of it.

That's why I think one of the most important questions in finance today isn't:

What is the next great fintech app?

It's: What infrastructure will the next generation of financial companies be built on?

Some of these categories will evolve dramatically. Some technologies will fail. New ones will emerge, but the direction seems increasingly clear.

The financial system is becoming more programmable and if that continues, the biggest opportunity may not be another application sitting on top of the existing financial system.

It may be rebuilding the infrastructure underneath it.

Last week I wrote that regulatory clarity could unlock capital. The question now is where that capital gets deployed.

That's where I'm increasingly looking.

CIO in Beta

Exploring the future of finance through the lens of capital allocation, AI, Ethereum and venture creation.