In 2026, Ethereum will begin to balance at a higher level with ZK

2026 is a critical year for Ethereum. The original Ethereum proofs produced tiny zero-knowledge (ZK) proofs instead of re-executing transactions. This immediately unlocks the benefits of Layer 1 and sets it at 10,000 transactions per second (TPS).

Researcher Justin Drake demonstrated that verifying proofs on an old laptop is already possible at DevConnector's EthProofs Day in November. One in 10 verifiers is expected to convert to ZK before the end of the year.

It's a complete reinvention of the fundamental way blockchain works: comparable to the merger in 2022, when Ethereum successfully transitions from proof-of-work to proof-of-stake.

Currently, each validator re-executes each transaction, and introducing a fully parallel process in Glumsterdam makes the process more efficient, just one step down the road.

The plan is to generate a ZK-proof for each block — a magic math that proves the block was actually executed — and then validators simply verify that the proof is legitimate.

This solves the blockchain trilemma nicely because ZK-proofing is so easy that in theory you can do it on your smartphone or even a smartwatch. This ensures that the network can remain highly decentralized without being burdened by very low specialized devices in the network. Ethereum can currently manage about 30 TPS (currently less), but the requirements for home verification are already at the level of a gaming laptop.

“It's a way to measure the network and measure traffic,” said Gary Schulte, senior staff member, blockchain protocol engineer at Besu Client. He explained that the gas limit cannot be raised much without adding minimal specifications for the authentication hardware. But with the new system, most of the hard work required for livestock equipment is done by block builders and ZK provers.

“If we have a handful of machines building these blocks, executing these blocks and validating them, and all of the downstream validator network is doing a much simpler job … it allows us to scale,” Schulte said.

Drake expects 10% of the network to convert ZK-proofs to Lean Performance's “Level 1” component this year. Since the validators most likely to switch are the lowest house validators, this allows the gas limit to increase, as the remaining validators who are still making transactions are running more capable machines.

“Sophisticated operators with large infrastructure will continue to be the future, and that means gas limits can be increased, because you don't have to worry about monopolies,” Drake told Bankless.

That said, the switch won't happen until mid-year, when Ethereum stops penalizing validators for late executions. This currently undermines the validity of ZK-proofs as these take a long time to generate and distribute through the network.

Schulte says that only verifiers “willing to pay fines” verify ZK-proofs first, but the ePBS amendment in the Glamsterdam amendment fixes the issue.

“Essentially, when the ban comes, you have a lot of time instead of witnessing immediately. You think about a whole slot to witness,” Drake said.

“And at that point, I would expect the number of verifiers to go from 1% to something closer to 10% opting in.”

How the release of Ethereum ZK is expected to work

Back in July, Sophia Gold wrote a blog post for the Ethereum Foundation outlining an ambitious goal of shipping zkEVM within a year.

Several groups have demonstrated the ability to generate proofs in real-time, which is actually two seconds faster than Ethereum's 12-second block times, given the need to distribute the proofs across the network.

Since all current systems have some errors and sometimes fail, a proposed solution provides ZK-proofs using five different verification systems, and as long as the verifier receives three related verifications, the block is good to go.

The goal is to eventually have a proof (proof of inclusion), but that generation system needs to be publicly proven to have zero errors. There's a team working on that, but it's not expected to be finished until 2030.

Verification doesn't need to be anywhere near as decentralized as accuracy (whether the proof is correct or not), but the target specifications are still set at the level of a device a home enthusiast can afford to run in their garage: one that costs less than $100,000 and requires the same amount of power as a Tesla Powerwall.

Given the impressive speed of development in the field, these specifications may be more than desirable: in May, SP1 Hypercube's 160 GPUs took less than 12 seconds to verify blocks. The ZKsync team can now verify blocks in 7.4 seconds using 24 GPUs, while the ZKsync Airbender team can verify Ethereum blocks in less than 50 seconds using one GPU, albeit with lower security.

Ethereum to ZK transfer: steps 0, 1 and 2

We're currently at phase zero of the rotation, where only fans like Drake are willing to take the plunge to prove it. Phase one is expected in 2026, with up to 10% of verifiers exchanged. Drake hopes season two will happen in 2027.

“Phase two is where a lot of the magic really happens, which is the mandatory verifications for the block producer to generate evidence and everyone is expected to work on the ZK EVMs,” he explained.

Expect to see more debate later in the year on whether the Ethereum Virtual Machine (EVM) should be converted to RISC-V (Reduced Instruction Set Five) to make it easier to generate ZK-proofs.

Related: Vitalik Buterin presents ZK-proofs for checking the X algorithm and steps

“The speed is with the RISC-V architecture because it's an open standard for licensing and royalty reasons and things like that, so it's a much smaller set of strict guidelines,” Schultz explained.

But at Devconnect, Schulte observed “a split among major devs about the feasibility and risks of ZKVM's scaling approach.”

“The concern I have heard is that the most experienced groups and software foundations are facing significant challenges in the zkEVM environment. Only relatively newcomers can compile to RISC-V locally, and most of the ZKVM groups are running their own modified version of RETH,” he said.

“I don't think it challenges the general direction of verification as a matching strategy, but rather the limitations that you believe real-time verification is too hard on the performing client software(s).”

Alex Gluchowski, co-founder and CEO of Matter Labs, told Cointelegraph that embracing ZK certifications will increase interoperability between chains using proven standards in the short term and “make communication more native” in the long term.

ETH in 2026: Ethereum's interoperability layer

Another big update for the Ethereum ecosystem in 2026 is the Ethereum Interaction Layer (EIL). It's a trustless messaging system that allows different Layer 2s to communicate and communicate, says Tyco Chief Operating Officer Joaquin Mendes.

“Ethereum now has 55+ L2 coils that have successfully scaled the network but created quiet ecosystems with fragmented liquidity. EEL will unify these into what looks like a single chain,” he told Cointelegraph.

Once it's up and running, a user can send 100 USDC from their Arbitrum wallet directly to another user's wallet, Base, which will receive it in seconds. Wallets can collect some Ether (ETH) from the roll address and some ETH from the Arbitrum address to pay for the invulnerable token on Linea.

Built using the ERC-4337 account abstraction, EEL is an attempt to improve solver networks like Intents, where users specify what they want to happen and how the solver network works with payments.

Mendes says that current solution-finding networks are insufficiently decentralized and require too much trust. “When a few major players dominate the market, competition decreases and controversial behavior increases. Without accountability frameworks and decentralization, solvers create censorship vectors that undermine Web3 principles,” he said.

“Idea-based systems rely on capital up front and solvers fulfilling transactions quickly. If there aren't enough solvers to compete, or if the cores merge, users lose competitive pricing and best performance benefits. The system only works well when there's a healthy and diverse solver network.”

EIL is specifically designed to enable wallets to perform cross-chain transactions without relying on intermediaries who steal or freeze funds or launch sandwich attacks. Instead of conducting transactions, liquidity providers are encouraged to withdraw funds in cross-chain pools.

“EIL is an account-based interop: the user's account makes calls directly on each chain. Liquid suppliers only supply gas and assets – they never enter transactions and never see call targets,” explains Yoav Weiss, developer of EIL. This eliminates the reliance on a ‘middle state' trust where a third party finder/forwarder transacts on behalf of the user.

Ethereum Foundation's Three-Phase Release Although it is targeting completion by 2026, it appears to still be in the research and development stage. The system will not perform as well as it should until fast L1 validations arrive at a future hard fork. However, the Open Intents Framework product contracts are already live.

It is a Taiko based package. It uses Ethereum validators for serialization, which provides the possibility of synchronous integration: real-time interaction with other established scrolls. “It's still a big milestone ahead, but it's closer than ever,” Mendes said.

“EEL also uses embedded wrappers. While embedded wrappers can easily write to each other in a common L1 sequence, EEL provides a non-standard messaging layer to seamlessly interoperate with embedded wrappers, improving connectivity across the wider ecosystem.”

Improving ZKsync's Atlas

Ethereum L2s are also using ZK-proofs. AppChain Lighter is using custom ZK circuits to scale up to tens of thousands of TPS and using ZKsync technology to create fast interactions between Ethereum L1 and ZKsync chains using the Atlas update and new gateway architecture. The first Atlas chain in production, UAE institutional chain ADI, has just gone live.

According to Gluchowski from Meter Labs, the update will allow coins to remain on the mainnet but be traded in ZKsync's Elastic Network for faster on-chain execution.

He explained that funds can be stored on Ethereum L1 but will be used immediately on L2s. At a higher level, “assets are held on Ethereum L1, and ZK-proofs are used to safely reflect their activity on ZK Chains.”

“From a user or application perspective, L1 origin assets on ZKsync chains are effectively real-time and inheriting Ethereum's security and finality. For institutional and RWA flows that already maintain Ethereum's level of finality, Atlas makes interrupt latency a bottleneck no longer.”

Atlas effectively means L2s can directly tap Ethereum's massive TVL, most of which have been reluctant to link funds to L2 until today.

“Historically, each L2 had to match its own liquidity, run different bridges, and rebuild the same DeFi stack, which is capital-inefficient and risky. With Atlas and Gateway, ZKsync chains can tap Ethereum liquidity directly in real-time instead of splitting it into separate pools. A single Ethereum block.”

Magazine: The Big Questions: Will Bitcoin Survive a 10-Year Blackout?