Vitalik Buterin Outlined a New Plan to Simplify Ethereum

Vitalik Buterin proposed a new plan to simplify the Ethereum protocol. He shared this in a blog post titled “Simplifying the L1”

In a recent blog post titled "Simplifying the L1," Vitalik Buterin, a co-founder of Ethereum, has proposed a new plan to simplify the protocol.

The post, which was published on May 3, 2025, sees Buterin suggest that Ethereum should aim to become "close to as simple as Bitcoin" within five years.

This move is intended to help reduce costs and increase security. Buterin admitted that some earlier decisions, including his own, have increased technical complexity and created inefficiencies.

"Historically, Ethereum has often not done this (sometimes because of my own decisions). We decided to add more technical DEFIgners to the sync committee system, we've kept the legacy EVM despite its flaws because it was deemed "good enough," and we've generally erred on the side of more features and more complex systems, even when simpler alternatives existed," he wrote.

Buterin added that the overengineering had led to delays and an increased risk of bugs.

"Whenever a technical DEFIgner is working on something, they can feel in their bones how much simpler and tidier their own internal mental model for the system is compared to the actual code that gets output, and how much simpler and tidier their own mental model is compared to the average person's understanding of the system," he explained. "We're also used to simpler code. For example, the Tinygrad project aims to build a deep learning framework that can fit entirely in a single C++ header file, and projects like LUAA attempt to make it possible to write entire AAA games in less code than the average mobile game."

3-Slot Finality Proposed to Simplify Ethereum Consensus

Buterin proposed 3-slot finality to simplify the Ethereum consensus process. This change would remove existing elements like epochs, sync committees, and validator shuffling.

3-slot finality reduces the confirmation process to three voting rounds within three slots, where validators first propose a block, then vote on it twice to confirm and finalize it. This model removes complex elements like epochs, sync committees, and validator shuffling, aiming to lower protocol complexity, improve safety, and allow for faster and clearer block finality across the Ethereum network.

"The reduced number of active validators at a time means that it becomes safer to use simpler implementations of the fork choice rule, and we can also simply have fewer types of messages that validators need to process and output (for example, no slashing messages). In general, we want to be able to reach a point where the total code size for the consensus-critical parts of the client software is no larger than, say, 100,000 lines of code, or ideally even less than 50,000 lines of code," Buterin added.

This proposal also suggests using STARK-based aggregation protocols. These are cryptographic systems that allow validators to prove and bundle large amounts of data, such as votes or transactions, into a single proof that others can quickly verify. In Ethereum, these protocols let any participant act as an aggregator without needing trust-based roles or preassigned committees. By removing sync committees and relying on STARKs for coordination, Ethereum can achieve decentralized, efficient validation with reduced overhead, making the network easier to scale while keeping it secure.

"We can also use STARK-based aggregation protocols to aggregate attestations and other messages, and crucially, these protocols can be implemented in a modular fashion. For example, we can have one open-source reference implementation of the STARK system, but different client teams can choose to use their own preferred programming language and build their own implementations of the STARK aggregator and generate proofs in a format of their choosing, as long as they can be checked by the reference implementation," Buterin explained.

Buterin proposes changing the Ethereum execution layer. He suggested replacing the Ethereum Virtual Machine (EVM) with RISC-V, which is a minimal instruction set architecture. RISC-V may help improve performance in zero-knowledge proofs, which are used in Ethereum’s privacy and scalability features.

The Ethereum Virtual Machine (EVM) is the core component that executes smart contracts on Ethereum. It was designed specifically for the Ethereum network and uses its own set of instructions tailored to blockchain logic. The EVM supports features like account storage, gas metering, and deterministic execution, but its complexity has grown over time. This makes it harder to verify with cryptographic proofs, especially for advanced systems like zero-knowledge proofs.

RISC-V, on the other hand, is a general-purpose, open-source instruction set architecture (ISA) commonly used in computer processors. It follows a minimalist approach with a small, consistent set of instructions, making it easier to implement and verify. Vitalik Buterin proposed using a ZK-friendly virtual machine based on RISC-V to simplify Ethereum’s execution layer. This shift could support faster zero-knowledge proof generation and reduce the overall complexity of the protocol