Author: Spencer Bogart, General Partner at Blockchain Capital; Translated and edited by Hu Tao, ChainCatcher.

Most people view onchain technology as a faster, more efficient version of existing technologies: faster payments, lower settlement costs, and more efficient capital markets. They’re not wrong. This alone represents enormous opportunity and will spawn numerous venture-scale outcomes over the next decade. But I believe this is only the smaller part of the story. When I look at this technology—and at the possibilities unlocked by programmable assets in a global, composable, always-on environment—I think we’ve barely scratched the surface. The most astonishing things have yet to be built. And they haven’t been built—not because the technology isn’t ready, but because we haven’t yet imagined them.

When the internet first emerged, its most obvious use case was communication. Email was faster and cheaper than physical mail. Email mattered profoundly—but its purpose wasn’t to speed up the post office. It stood on its own and rapidly proliferated. So if you’d assessed the internet in 1995 and observed widespread email adoption, you could reasonably conclude that earlier theories had been validated. Yet most of the opportunities hadn’t even begun to germinate. Search, social networks, e-commerce, cloud computing, software-as-a-service (SaaS), and streaming weren’t “accelerated versions of existing things”—they were entirely new categories that couldn’t possibly exist before the internet created the necessary conditions. Google isn’t a faster library; Facebook isn’t a faster phonebook; AWS isn’t a faster server room. They only make sense after a globally connected, programmable network exists. Collectively, these new categories dwarfed the “faster communication” use case by several orders of magnitude.

I believe crypto is now in its explosive growth phase. Most attention focuses on how to run existing financial products better onchain—faster settlement, cheaper cross-border payments, tokenized Treasuries and equities, and more efficient lending markets. And these efforts are bearing fruit: stablecoin settlement volume is projected to reach $33 trillion by 2025, and the market cap of tokenized Treasuries has recently surpassed $15 billion. The world’s largest asset managers and banks are building businesses on public blockchains. That’s fantastic. I’m excited about all of it—and I spend my days working on it. But this is just the most obvious application, one that fits perfectly with our existing knowledge frameworks and is so large it can easily be mistaken for the entire opportunity. The question I find far more interesting is: What becomes possible only when programmable resources exist in a global, composable, always-on, permissionless environment? What new verbs emerge? What categories remain unnamed?

We already have at least one clear example worth studying carefully—because it illustrates a pattern I expect we’ll see repeatedly. Imagine being able to borrow $1 billion without collateral—and with mathematical certainty that the lender will be repaid. How would you feel? That’s a flash loan: borrowing any amount of funds without collateral, provided the loan is repaid within the same transaction. If repayment fails, the entire transaction automatically reverts—as if it never happened. Lenders bear zero risk. No credit checks. No relationship-building. No collateral. Just system logic providing the guarantee. Before flash loans existed, no one needed them. Why? The concept is completely alien to traditional finance. Even before programmable assets existed, it was useless—so there was no existing category to improve upon. Uncollateralized, unlimited, mathematically guaranteed lending is structurally impossible in any system where transactions require time to settle. It only becomes possible when execution is atomic, assets are programmable, and the entire sequence of operations either completes fully—or doesn’t happen at all.

Once atomicity makes it feasible, flash loans become a standard tool in the onchain economy—used for arbitrage, liquidations, collateral swaps, and capital-efficiency strategies that simply cannot exist in traditional payment systems. Of course, any powerful new technology is inevitably misused—and that very misuse highlights the novelty of its underlying mechanics. Flash loans didn’t make lending faster or cheaper. They created a new kind of lending—one that was structurally impossible before programmable assets and atomic execution existed. This is what I mean by a “new verb” or “new action.” The system can now do something it fundamentally could not do before—not because someone optimized an existing process, but because the underlying primitives themselves changed.

But I must confront the limits of my own imagination honestly. I can describe this design space in abstract terms. Public blockchains introduce a set of foundational primitives that previously did not exist: atomic execution, shared global state, programmable custody, deterministic settlement, composability across independent participants, and software-native assets. We’ve never had a financial system that integrates settlement, custody, liquidation, and execution—all within a single programmable environment. When layers that used to be separate fuse into one, new things become possible. But I cannot tell you precisely what those things are. And I think that’s exactly the point. Human imagination is backward-looking. We’re exceptionally good at improving on what already exists—but much less skilled at imagining things that were literally impossible yesterday. We look at onchain technology and instinctively ask: What existing products can it make faster and cheaper? The harder—and more valuable—question is: What entirely unprecedented things can it create?

I have some intuitions. Programmable custody systems that enforce complex protocols without intermediaries. Capital entrusted to software agents operating within bounded parameters. Financial structures dynamically assembled and dissolved in real time based on onchain-verified conditions. These directions feel right. But the most important applications may be ones I cannot yet describe—because they’re unlike anything I’ve ever seen before. The fact that they resist enumeration is, in fact, the strongest evidence for the argument: If I could easily list all the truly new things, they wouldn’t be truly new. The design space is vast—most of it remains unexplored—and intuition alone cannot map it. That’s the crux.

So most attempts in this domain will fail. A vast design space doesn’t imply easy outcomes. But the opportunities embedded in the few approaches that do work are enormous—and for the past thirteen years, we’ve been building pattern-recognition capabilities precisely to spot them before they become obvious. It’s this opportunity that makes me so excited about the next decade. Most of the opportunity still lies ahead. If the internet analogy holds, then the equivalents of search, social, cloud computing, and SaaS—within the onchain economy—have yet to be built. Email was a trillion-dollar industry—and the services built atop it were worth many trillions more. I believe that ten years from now, looking back, what will excite us most will be things that don’t yet exist. This isn’t just about making banks, exchanges, or asset managers more efficient—it’s about things that are possible only in a composable, global, 24/7 environment with programmable assets. These things will seem obvious in hindsight—but today, we can’t foresee them, because they have no precedent. Flash loans give us a glimpse—but they’re just the tip of the iceberg. The design space is immense—and we’ve only just begun exploring it.

[ChainCatcher]