Signal founder claims that Ethereum has already failed—did he get it right?

Author: Jason Chaskin Translation: Shan Oba, Golden Finance

If Ethereum’s core value is to build a credible, neutral base layer that eliminates intermediaries, has it fulfilled that promise now? After more than ten years of development, Ethereum is already the public chain with the highest level of decentralization worldwide, the strongest neutrality, and the best security score. In theory, it should have long since realized its original vision.

The importance of this question has already driven the Ethereum Foundation to complete a reorganization of its internal structure.

Let’s turn the clock back about four and a half years. Signal founder Moxie Marlinspike raised an identical question and reached the conclusion: Ethereum is heading toward failure. He found that, for ordinary users to use Ethereum end-to-end, almost every step depends on centralized intermediaries. Web3 was supposed to decentralize the internet again, but Ethereum is repeating Web1’s old mistake—traffic and resources keep concentrating in large platforms. He wrote: “To make a technology like Ethereum usable, the entire industry is once again moving toward platform monopolies.”

“Usable,” as Marlinspike put it, theoretically does not require any trusted third party: users only need to run their own nodes to interact directly with Ethereum. But in reality, almost no one is willing to do so. Just like in the Web1 era, users didn’t want to build their own servers and instead leaned on major internet companies; today, users also don’t want to maintain blockchain nodes, so they have to rely on third-party service providers.

To support his viewpoint, Marlinspike developed a set of NFT applications to fully replicate the user operation flow. Users download a wallet and connect to the application; the first step in the wallet is to query real-time on-chain status from RPC providers such as Infura and Alchemy, and the entire wallet indiscriminately trusts the data returned by them. When users sign and mint NFT transactions, the broadcasting channel is still these centralized service providers. After the transaction is completed, the wallet calls the Etherscan interface to read transaction records, again unconditionally trusting the third party. And to display newly minted NFTs, it also needs to call OpenSea’s data interface. Even the front-end page is simply a standard React website deployed on a centralized server.

After using an Ethereum application end-to-end, the intermediaries users need to trust include: RPC node service providers (Infura/Alchemy), block explorers (Etherscan), NFT data platforms (OpenSea), and centralized website servers. It’s called a permissionless system that supposedly doesn’t require trust, yet it stacks layer upon layer of third parties that users must hand over trust to.

Marlinspike’s final judgment: “I don’t think Ethereum can lead us out of the stranglehold of centralized platforms.”

Ethereum founder Vitalik then published a post in response. He agreed with Marlinspike’s description of the ecosystem’s current state, but did not agree with his pessimistic conclusion.

Vitalik said Ethereum had already planned a user-experience roadmap that minimizes trust, but the prerequisite is completing a series of major protocol upgrades at the underlying layer—most importantly, moving from proof of work (PoW) to proof of stake (PoS). Once the upgrades are implemented, developers can build tools for autonomous verification like light clients. Wallets and applications no longer need to blindly follow centralized service providers; they can independently verify on-chain data. Developers can then package these verification capabilities into toolkits and SDKs, allowing all kinds of applications to connect directly. Ordinary users wouldn’t need to run full nodes and could still achieve low-trust interactions.

At the end of his post, Vitalik made a prediction: “A decentralized blockchain ecosystem with comprehensive autonomous verification capabilities is coming soon—far closer to reality than most people think.”

Four and a half years have passed. How should we judge Marlinspike’s criticism back then?

To this day, ordinary users’ usage path is no different from what Marlinspike described: throughout, it relies on centralized RPC service providers, centralized front-ends, and various third-party platforms. Objectively speaking, Vitalik’s prediction back then was overly optimistic—the ideal low-trust ecosystem still hasn’t taken shape.

But that doesn’t mean Marlinspike’s assessment of Ethereum’s final direction is correct. The Ethereum Foundation’s recent organizational restructuring actually confirms this.

Ethereum’s grand goal is to support global-scale, sustainable autonomous computation, but engineering resources are not infinite, and it can’t solve all problems at once.

In the years after Marlinspike published his post, the industry focus was entirely on building foundational infrastructure at the protocol layer: Ethereum successfully transitioned from PoW to PoS, which was one of the largest upgrades in the history of existing major distributed systems, laying the groundwork for all subsequent optimizations. Scaling is another big mountain. When Marlinspike wrote at the time, minting NFTs often required paying hundreds of dollars in Gas fees; today, Ethereum’s ecosystem processes transaction volumes up by several orders of magnitude, and scaling iteration has never stopped.

With the foundational infrastructure largely in place, the industry has finally refocused on the core pain point Marlinspike directly pointed out—the problem of the rampant proliferation of centralized intermediaries.

Now we’re at the present. A few weeks ago, the Ethereum Foundation announced an internal reorganization, and the access layer became the second-largest business segment. The core goals of this segment are: to ensure that when users call upper-layer applications, Ethereum’s core values of decentralization, neutrality, and security won’t disappear out of thin air, while making privacy protection a foundational capability. Solving the rampant proliferation of centralized intermediaries is precisely the core work of the access layer.

I myself will also take part in research and development related to the access layer. In the future, I’ll publish more articles to explain in detail how this segment can address the industry’s problems from back then. Being able to help narrow the gap between Ethereum’s vision and users’ real experience is something I’m truly looking forward to.