Who Can Make Money in the Age of Agents?
TechFlow Selected TechFlow Selected
Who Can Make Money in the Age of Agents?
The next billion users will be agents—and the crypto world hasn’t yet found their wallets.
Navigating Web3 tides with focused insightsAuthor: Jonah Burian
Translated by: Jiahuan, ChainCatcher
Many speculate that the next billion users of blockchain will be Agents. But few ask the deeper question: Who will actually make money in that world?
Every prior theory of value capture in crypto assumed users were human. The “Fat Protocols” thesis posited that protocols are best positioned to monetize human users.
By contrast, the “Fat App” thesis—explored by me and colleagues in How to Capture Value and The Great Re-rating—argues that the application layer can do it better. Yet Agents fundamentally change the nature of the user—and thereby invalidate both existing theories.
The Fat Protocols Thesis
In 2016, @jmonegro introduced the “Fat Protocols” thesis. For nearly a decade, it has been the dominant value-capture framework in crypto.
Its core claim is that in traditional internet markets, value accrues to the application layer (@Google, @facebook), while underlying protocols (TCP/IP, HTTP) capture almost none. Crypto would reverse this entirely. Because blockchains publicly share data, applications would gradually commoditize.
And because using the network requires consuming protocol tokens, those tokens capture speculative value as usage grows. Every successful app increases demand for the token. The underlying protocol’s growth rate would outpace any app built atop it.
For years, this seemed correct: Bitcoin and Ethereum’s valuations far exceeded those of any company built on top of them.
This model works perfectly when protocols themselves are scarce, costly to build, and difficult to replace. Bitcoin and Ethereum in 2017 were indeed scarce—there weren’t dozens of general-purpose L1s competing for the same workloads.
Blockspace was sufficiently constrained that holding the base asset felt like owning a piece of every application needing that asset.
Today, reliable alternatives exist at every layer of the infrastructure stack: multiple high-throughput L1s, dozens of L2s, and fiercely price-competitive modular settlement and data availability (DA) layers. Blockspace has shifted from constrained to abundant.
As cross-chain bridges and aggregators render underlying chains nearly invisible to users, switching costs have collapsed. Infrastructure has become interchangeable—and interchangeable commodities compete solely on price. As scarcity vanishes, so does protocol pricing power.
The Fat Apps Thesis
By 2026, the entities capturing most economic value will be apps—not protocols: e.g., @phantom, @coinbase, @Polymarket, @Pumpfun.
In my view, the most valuable asset in crypto is user relationships.
If you control the user interface and transaction flow, you control distribution—and thus profit from nearly any onchain product users interact with: swaps, lending, staking, minting, fiat onramps. This may also explain why funds are so obsessed with neobanks.
Apps also push infrastructure into pure price wars, compressing infrastructure margins down to marginal cost. I documented this dynamic in “How to Capture Value.” The same pattern is now playing out in stablecoins—a topic I’ve explored elsewhere.
Asset prices reflect this thesis. Spencer and I call this shift “The Great Re-rating”: in this cycle, value begins migrating toward the layer that owns users.
Why Agents Break This Logic
The Fat Apps thesis assumes users are humans who care about UX, brand, and convenience. But Agents don’t care about any of those. They call APIs directly, have zero brand loyalty, and switch platforms at zero cost.
When users become software, controlling user relationships ceases to be an unassailable moat. The entire frontend moat upon which the Fat Apps thesis rests is failing.
So who captures value in the Agents era?
Apps Go “Headless”
In one plausible future, winners at the application layer retain dominance by shedding their frontend interfaces—going “headless.”
Wallets and aggregators have already done the hardest engineering work: integrating with dozens of protocols, building routing logic, authentication, and fiat onramp infrastructure.
The natural next step is to expose this entire stack as an API for Agents—letting Agents route through them just as humans currently route through @phantom or @JupiterExchange.
In this world, the Fat Apps thesis survives—but loses its frontend. Companies that won in the human era will transform into pure backend infrastructure for Agents. We’re already seeing traditional SaaS firms like Salesforce move in this direction.
Protocols’ Resurgence
In another scenario, Agents bypass intermediaries entirely.
If integration becomes simple enough (well-documented APIs, standardized RPCs, predictable execution semantics), Agents have no real reason to pay aggregators to do what they can do themselves. Aggregators’ human-era advantage was UX and managing routing complexity.
But Agents need no UX—and routing is an engineering problem increasingly within Agents’ capabilities.
If the world evolves this way, the Fat Protocols thesis enjoys a second wind.
Pricing Power Collapses Across the Stack
Perhaps Agents apply commoditization pressure everywhere. They are perfectly rational, routing frictionlessly and disloyally to the cheapest trading venue every time.
Apps lose the ability to charge UX premiums to humans. Aggregators and infrastructure alike lose pricing power—because there’s no longer human inertia shielding them from price wars.
In this scenario, no layer captures much profit. Margins across the entire supply chain compress toward marginal cost—and residual value flows to Agents’ owners—or to the end users Agents serve.
Crypto becomes a utility—and utilities rarely generate outsized profits.
Agents Unlock Unprecedented Activity
A simple take is that Agents perform all the tasks humans do—just faster and at greater scale. Even with compressed margins, the total pie still grows.
But I find a more interesting version compelling.
Agents enable activities previously infeasible: e.g., continuously rebalancing portfolios at sub-cent execution costs; machine-to-machine commerce between Agents; and entirely new markets made possible by pricing and trade speeds that exceed human reaction thresholds.
Current onchain activity metrics don’t reflect these yet—because we implicitly assume human involvement.
If this is indeed Agents’ transformative impact, the question shifts from “How will the existing pie be divided?” to “How much new economic activity will flood onchain—and which layers are best positioned to serve it?”
An Unnamed Business Model
In every cycle, we try to predict where value will flow—and tend to assume existing business models will persist. That assumption often blinds us to novel models not yet conceived.
At the dawn of the internet, no one foresaw the attention economy. The idea that “slicing user attention and auctioning it to advertisers would become the dominant business model—and that a single company could capture a significant share of global ad spend”—was utterly alien at the time. Only in hindsight did it seem inevitable.
AI appears to be the largest technological disruption in decades. In an Agents-dominated world, some value capture may flow to business models no one mentions today—and the entities capturing that value may not be the ones markets currently watch.
What to Watch
The most likely outcome isn’t one system fully replacing another. Humans and Agents will coexist as crypto users for a long time—and their respective value-capture landscapes will differ sharply.
So long as humans interact with chains, the Fat Apps thesis holds: consumers willing to pay for UX, brand, and convenience will continue to reward apps that own those relationships. Meanwhile, the layer touched by Agents—regardless of which of the above scenarios materializes—will operate under a separate, independent theory.
For builders, the recurring question on the Agents side is: What makes an Agent choose you—rather than route straight to the next cheapest alternative? UX is unlikely to be the answer. Liquidity, latency, settlement guarantees—those may be.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
@jonah_b
