The numbers do not lie, even when the narratives do. Over the past 90 days, I have tracked the launch of at least seven new Layer 2 networks, each with a total value locked below $50 million. Collectively, they represent less than 2% of the daily active addresses found on Arbitrum or Base. This is not scaling. This is slicing.
We have entered the era of the Layer 2 glutโdozens of rollups, each with their own token, their own governance forum, and their own ecosystem fund, all competing for a share of a user base that has barely moved in years. We have seen this movie before. In 2017, the ICO boom gave us hundreds of tokens promising to revolutionize every industry. 2021 gave us the app chain thesis. 2023 gave us the modular blockchain narrative. And now, in this quiet sideways market of 2025, we are being sold the Layer 2 ecosystem as the ultimate solution to Ethereum's scalability woes.
But here is the uncomfortable truth, based on my years watching decentralized protocol architectures evolve: we are not building more users; we are just building more ways to divide the existing ones.
Let me walk you through the data I have been tracking. In the last quarter, the total value locked across all Ethereum Layer 2s has remained stagnant at roughly $35 billion, even as the number of networks has grown by over 30%. The daily active addresses across these networks have not grown proportionally. Instead, we have seen a redistribution of the same users, the same capital, and the same liquidity farmers hopping from one incentive program to the next.
This is fragmentation disguised as innovation.
I remember a time when scaling meant something specific. It meant more transactions per second, lower fees, and a better user experience. It meant that a person in Southeast Asia with $50 could actually use a lending protocol without losing half their deposit to gas fees. It meant that the promise of permissionless access was not just a slogan, but a technical reality.
Now? Scaling means creating a new token so that a venture fund can mark a return on their spreadsheet.
Let me pause to issue a clarification. I am not against Layer 2 technology. Zero-knowledge rollups are a mathematical marvel. Optimistic rollups have an elegant mechanism for fraud proofs. The technology is sound. The problem is not the technology; it is the metaphysics of the market. We have confused the creation of a new network with the creation of new value.
There is a friction in the system, and not the kind you measure in gas costs. I am talking about the friction of liquidity dispersion. Each new rollup fragments the pool of available capital, making the entire ecosystem less efficient for traders, less liquid for borrowers, and more complex for the very users we claim to want to onboard. Every new bridge is a new attack surface. Every new canonical token standard is a new point of confusion. Every new sequencer set is a new vector for centralization.
But here is where the analysis gets uncomfortable for those of us who believe in this space.
In my early years, I believed that a permissionless network was the ultimate destination. I still do. But I have come to understand that permissionlessness is not a virtue; it is a property. It is a characteristic of the architecture, not a promise of success. Just because you CAN build a Layer 2 does not mean you SHOULD. Just because you CAN issue a token does not mean you HAVE created value.
We are seeing a proliferation of Layer 2s, not because users are demanding them, but because the economic incentives for founders and VCs are misaligned with the economic needs of the ecosystem. This is a structural problem, and structural problems require structural solutions.
What would a structural solution look like? It would require us to become the gatekeepers of our own attention, the validators of our own intent. It would require us to understand that the protocol remembers what the market forgets: that value accrues to networks that serve a real purpose, not to networks that simply occupy a slot in a multichain future that never arrives.
The contrarian view, of course, is that this fragmentation is a natural phase of growth. In the early days of the internet, we had WebVan and Pets.com, and out of that chaos came Amazon and Google. The same Darwinian process will apply to blockchains, and the best networks will rise to the top.
I was a champion of this view for a long time. I believed, with the earnestness of an idealist, that the market would eventually sort it out, that the invisible hand would guide us to a future where only the most robust networks would survive. But here is the problem with that Darwinian analogy: the cost of experimentation in the internet era was measured in dollars, while the cost of experimentation on a fragmented blockchain ecosystem is measured in diluted trust.
Every time a user gets stuck bridging assets to a chain with no liquidity, every time a wallet asks them to configure a custom RPC, every time they lose money to a bridge exploit because they sought yield on a network with $10 million TVL, the entire ecosystem pays a price. We are bleeding trust, not dollars. The gatekeepers of the traditional financial system watch this chaos and see no reason to enter. We are asking them to trust a world where our own technology cannot even keep its users on a single network.
I have a memory from 2020. I was working with two friends on modeling the impact of undercollateralized lending on underbanked populations in Southeast Asia. We would sit in a small apartment, night after night, poring over Compound's mechanics, running simulations, trying to figure out if this architecture could actually solve a real problem in the real world. The conclusion of that research was sobering: while technically efficient, the system still replicated traditional banking exclusion through over-collateralization. We built a beautiful engine, but we put the same old gatekeepers inside it.
That memory has informed everything I have written since. It taught me that a protocol does not exist in a vacuum; it exists in the context of the people who use it, and the incentives that shape their behavior. Now I sit in London, watching the industry repeat the same mistake. We have built a beautiful engine of Layer 2s, but we have failed to ask who this service is truly serving. Are we onboarding the underbanked? Are we creating permissionless markets for the unbanked? Or are we creating a series of rotating doors, designed to keep the same users, the same capital, and the same attention hopping from one incentive program to the next?
I am not naive. I understand the market dynamics that drive this. Venture capital funds require returns; they have a shelf life; they need to deploy capital. Founders need to raise; they need to show progress. All of these are reasonable incentives within the existing system. But if we are truly building for the long term, if we truly believe that decentralized networks can fundamentally reshape the power dynamics of the global economy, we need to resist the urge to confuse motion with progress.
Stillness would serve us better. Patience is the validator of true intent. Instead of building the 30th version of a general-purpose EVM rollup, we should be asking: what networks actually require their own execution layer? What problems are so unique that they cannot be served by the existing architecture? The answer, I suspect, is far fewer than the number of networks currently in development.
Let me make a prediction. In the next 18 to 24 months, we will see a consolidation. Not the kind that comes from acquisitions, but the kind that comes from neglect. As liquidity continues to fragment, the marginal Layer 2s will dieโnot with a bang, but with a whimper. Their tokens will quietly lose 90% of their value. Their Twitter accounts will stop posting. Their developer communities will drift to more vibrant ecosystems.
The infrastructure narrative will shift to something new; perhaps a focus on aggregation layers, attempting to solve the fragmentation problem with smart accounts and unified liquidity. But the core defect in our thinking remains unaddressed: we are still too focused on building new networks and not focused enough on building on the ones we already have.
I am an evangelist for decentralization, but I am also a pragmatist. I have watched too many dreams die in protocol whitepapers to simply cheer for the next launch. I have seen what happens when the market gets distracted by novelty. Consider how this pattern reads to the institutional world. In 2024, when the Spot Bitcoin ETF was approved, I was asked to consult for a major UK pension fund. We spent hours in rooms with traditional finance stakeholders, and they asked the same question over and over: "If I invest in this technology, how do I know you are not just going to build another shard or another rollup that makes my investment obsolete?"
The answer I gave was not what they expected. I told them that the technology itself was not the concern. The concern was whether the ecosystem could stop cannibalizing itself. I told them that Bitcoin was a neutral reserve asset, a store of value, but that the real promise of blockchains was in the application layer. And the application layer was in danger of becoming a museum of orphaned tokens.
Optimism is a trait I have in surplus, but it does not mean I am blind to the silence when the noise stops. In the current market, the noise has stopped. We are in a sideways market, which is the best time to observe the true signal of protocol quality. The hype has died down. The retail speculators have gone home. What remains, on each network, is a core of genuine believers and the actual use cases that persist.
Over the past 7 days, I was tracking the TVL of a well-known optimistic rollup. It lost 40% of its LPs in a week. The network is not new; it has been live for two years. It has a token. It has a governance forum. It has monthly upgrades. And it is losing liquidity because the market has realized that its token is only used for governance, and its governance does not matter, because the sequencer is still run by a foundation. And if the foundation is controlled by a board of directors, who is the protocol really serving?
The blockchain community is built on a promise of transparency, yet the very people building decentralized applications are often the most hesitant to share their data. If you build a network, show me the numbers. Show me the compounding growth of daily active users. Show me the daily transaction volume that is not 80% wash trading or self-transfers. Show me that my assets are safe on your bridge. Show me your code, your audit results, and your resistance to game theory exploits.
I use all this data as a sieve. If a network can survive my putting its claims through a filter of empirical analysis, I will listen. But so many networks cannot withstand this level of scrutiny.
I have been in this industry for a long time. I have written thousands of words about the promise of decentralized systems. I have felt the emotional peaks of seeing a protocol launch and the valleys of seeing it fail. I have sat in a cabin in the Scottish Highlands, processing the emotional toll of the Terra/Luna collapse and questioning why I still believe. And still, I believe. But my belief is not measured in blind faith; it is measured in daily steps. It is measured by the 20 lines of code a developer writes to fix a critical vulnerability in a smart contract, even though no one will give them an award for it. It is measured by the community member in Nairobi who decides to teach a DeFi course even though their own wallet holds less than $20. It is measured in small counts, not big claims.
As we look toward the coming year, our focus should not be on replicating the general-purpose machine. We should be looking for those who will do the unglamorous work of connecting the fragmented pieces through a seamless UX that abstracts away bridges, rollups, and gas fees. We should build in silence so the network can speak.
The next narrative should be one of convergence, not fragmentation. It should be one where multiple execution environments share a common settlement layer, and users do not even know which domain their transaction is being executed on. It should be one where the test of our success is not the number of tokens launched, but the number of active users who stay.
I remember a conversation I had in 2017, back when the ICO mania was peaking, and I was auditing the whitepaper of a pioneering decentralized exchange. I read about relayers and off-chain order books, and I realized that true freedom lay in permissionless access, not the permissioned liquidity being offered by the exchanges who were raising millions. That insight has shaped my entire career. It taught me that architecture matters more than asset price. That the things we build in silence will speak louder than the things we shout from the mountaintop. That the protocol remembers what the market forgets.
And the market is currently forgetting that we are building a parallel financial system. It is forgetting because we are too busy building parallel, fragmented trading environments and pretending that our ability to launch a chain on a rollup framework is the same as our ability to change the lives of the 1.4 billion people worldwide who remain unbanked.
Let me offer a new model of thought: we should think about Layer 2s not as destination networks but as infrastructure service layers. Think of it like a utility grid. In the real world, you do not have a separate grid for every neighborhood. You have one grid, with multiple points of access, all connected through standardized relays. The user does not have to think about which power plant generated their electricity; they just plug in their device and expect it to work.
That should be our model for blockchain. We should not be asking users to navigate a complex matrix of bridges, canonical token standards, and sequencer sets. We should be asking them to plug in and play, not because we have built yet another chain but because we have built better connections. When I look at the current landscape of fragmented Layer 2s, I am reminded of the early days of the AOL walled gardens. Back then, you had CompuServe, Prodigy, and AOL, each with proprietary email systems and content. They were scaled networks, but only within their own boundaries. And then came SMTP.
Now we are building the walled gardens of blockchain. Each Layer 2 is a garden with its own bridge, its own token standard, its own learning curve. But where is our SMTP? Where is the protocol that seamlessly allows cross-network messaging and value transfer without technical friction? We have attempted to solve this with interoperability protocols, but these have introduced an additional layer of security risk and are often too complex for mainstream users. We are still missing a standard user-facing interface. We are still missing a chain-agnostic way to hold assets.
But I sense the beginning of something else. The rise of smart accounts on ERC-4337, the maturation of intents and shared order flow, and the increasing ambition of solvers to abstract the UX of a decentralized exchange. This is a return to the original ethos: not to know what chain you are on, but to trust the intent of a user. I am starting to see the foundation of an interface where users do not need to know what a bridge is or how a rollup is named, because the system will abstract it for them. This is the kind of structural innovation that we should be writing about, not the launch of the 40th network.
The longer we are in this sideways market, the more time we have to build meaningful connections between fragmented networks. The longer we sit here in silence, the more time the market has to realize that Token X and Token Y on some new chain are not what matters. What matters is whether the underlying metadata can preserve the authenticity of human transactions for the next thousand years. What matters is that 90% of the current Layer 2 projects will eventually not survive, and their death will be brought on by liquidity drought and a lack of clear utility. They will be remembered as a footnote, like the altcoins in the 2018 era that claimed to be Ethereum killers.
What would be a test of a successful Layer 2? I propose a simple one: if a new user never interacts with a bridge in their first six months of using a network, then we have succeeded in the scaling mission. If a developer can deploy an application to the chain in minutes because it uses a standard framework, we have succeeded. If a user can access the application without having to pour gas tokens into a separate chain and then a separate rollup and then a separate bridge, we have succeeded.
I have been tracking this niche with the care of a natural historian, documenting frameworks such as the OP Stack, Arbitrum Orbit, and the various zkEVMs. These are not just tools for spinning up a chain; they are templates that allow founders to launch a governance token within hours. But then the question remains: what do these chains do? Let me take a step back to get the wider view. Sooner or later, the market will realize that the Layer 2 story is not a story of a hundred chains but of a handful of networks that provide the best security and the best user experience. The next step is to integrate those few into a seamless product for end users. Instead of focusing on who is next, let us focus on what is next.
And what is next? AI agents are increasingly becoming a new category of market participants, and they need cryptographic credentials to transact on our behalf. We need to build network infrastructure not just for humans but for the agentic economy. That is what I am watching. That is what I find genuinely exciting. This is not hype; it is an infrastructural need. Machine-to-machine payments on open rails, micro-transactions at a level that no traditional payment system could handle, and provenance tracking to ensure that we know which content is generated by a person and which is generated by a machine. Here, blockchain has a genuinely unique role to play.
It is precisely in this context that we need to maintain the integrity of the cryptographic layer. We should not let Layer 2 fragmentation soil the foundational principle that made this space worth building in the first place. Trust is not given; it is verified. And when we look at all these networks, we need to verify their numbers and verify their intent. We need to be less forgiving of vanity metrics and more demanding of transparent accounting.
And so let us go back to where I started: with the user and the protocol as the unit of analysis. I am not discouraged by the market. Sideways markets are for positioning, not for panic. But we are not observing this market from the outside; we are inside the protocol, and we are building toward a horizon that is much farther away than we often contemplate. The next six to twelve months will reveal which Layer 2s are building infrastructure and which are building sandcastles.
When the current waves of networks reset, the chains that remain will be those that have the best economic alignment with their communities. Their tokens will be actively utilized for more than just governance theater; their venture funds will unlock and prove their conviction; their user bases will be real people, not incentive farmers. Code is the only permission we truly need. We have the tools. We have the frameworks. We have the funding. What we lack is the discipline to resist the urge to build a new island and the courage to deepen the channels of the ones we have.
Ask not what new chain can be launched in this cycle. Ask what existing infrastructure needs to be deepened, secured, and made accessible. The user is waiting. The protocol is ready. And we, with our hands on the keyboard, must choose to be the builders of connections, not the miners of fragmentation. The market is quiet. The silence is an opportunity. The network is fragmented. The integration is the mandate. This is not a call for the abandonment of innovation but a call for the highest form of innovation: making the complex appear simple, making the fragmented feel whole, making the technology invisible so that the value can become visible.
We do this not for the speculative speculator, but for the everyday user who simply wants to access a fairer financial system without having to navigate a maze of bridges and optimistic proofs. In this quiet market, surrounded by the eerie silence of unverified networks and losing LPs, I find my signal and my calling. Unite the pieces, build the connections, and let the network speak for the value that we have coded into its core. Code is the only permission we truly need, and trust is not given; it is verified. So let us verify the intent, choose the path of integration, and in our silence, let the network broadcast at full volume.
The next big chapter of this industry will not be written in gas fee comparisons of Layer 2s but in quiet moments when a developer decides to write a standardized interface that works across every chain. That is the quiet work. That is the invisible integrity that keeps the promise of blockchain alive. Take your positions now, not for market speculation, but for infrastructural certainty. The chop is just the calm before the integration. And the integration will be glorious.