Product

ZK Infrastructure Update: Transitioning from Proof Verification Layer to Proof Aggregation

TL;DR

Zero-knowledge (ZK) technology has matured tremendously since we started building Aligned, and Ethereum itself is moving toward native ZK support at the protocol level. We remain more optimistic than ever about the future of ZK. Everything we set out to accelerate is now happening. In this context, our Proof Aggregation Service, which inherits full Ethereum-level security, serves our clients' verification needs and is where our focus lies going forward.

Against this shifting landscape, we are deprecating our Proof Verification Layer. In coordination with our operators, tasks are no longer being sent to it, and they have been invited to deregister and shut down the instances currently running them as soon as possible. Our explorer will be updated to focus exclusively on the Proof Aggregation Service.

Introduction

Zero-knowledge has changed dramatically since we started building Aligned, and proof aggregation with full Ethereum security is now enough to deliver what we set out to achieve.

When we proposed Aligned more than two years ago, zero-knowledge adoption was being held back by a very specific set of problems: high verification costs on Ethereum, low throughput, and high latency. We set out to build a decentralized ZK verification layer that would remove those bottlenecks and make ZK the default for building provable, secure applications.

In the months that followed we shipped multiple testnets, including a private testnet built using the Cosmos SDK. In June 2024 we launched our first public testnet, opening our first EigenLayer (now known as EigenCloud) AVS to external operators. In November 2024, after nine months of development, we launched the mainnet beta of Aligned's ZK verification layer on Ethereum, going live with over 400K ETH securing the network so partners could start building immediately.

Since then, the product has done what we built it to do. But the world around it has changed, and this changes what makes sense for us as a next step.

Today we are announcing that Aligned's Proof Verification Layer will be deprecated, and that we will establish our Proof Aggregation Service as the default and only solution going forward.

Why we are deprecating the Proof Verification Layer

Three things have shifted since we designed the product, and together they make the case clear.

1. ZK has matured considerably since we conceived the product. When we started, fast and affordable verification was a bottleneck that required a dedicated layer to solve. Since then, proving systems have improved dramatically, and proof aggregation with full economic security from Ethereum is already a reality. The solution we built to reduce Ethereum's verification cost is less necessary now than it used to be, because the underlying technology got better.

2. Ethereum is moving toward enshrining ZK. The roadmap is pointing clearly in the direction of native ZK support at the protocol level. Building a verification layer that sits between developers and Ethereum makes less sense when Ethereum itself is going to handle this natively in the medium term. The right move is to align with where the ecosystem is going. Proof aggregation with full Ethereum security fits that direction; a dedicated verification layer no longer does.

3. The market prioritized full Ethereum security over ultra-low latency. One of the strongest arguments for a dedicated fast-mode verification layer was the assumption that applications would need sub-block latency for ZK proofs. What we saw in practice is that developers cared more about inheriting Ethereum's security guarantees than about shaving off latency. Our Proof Aggregation Service delivers exactly that, and it is enough to meet what our clients need.

What the Proof Verification Layer accomplished

Before moving on, it is worth stating clearly what this product actually did.

Aligned's Proof Verification Layer verified more than 136,000 proofs on Ethereum. 59 operators were active on the network at some point during its lifetime. The network supported a wide range of proof systems, giving developers flexibility that was not available anywhere else:

gnark: Groth16 (with BN254) (v0.12.0)
gnark: Plonk (with BN254 and BLS12-381) (v0.12.0)
SP1 (v5.0.0)
Risc0 (v3.0.3)
Circom (v2.2.2)
Mina (testnet only - Sepolia, Hoodi)
Mina Account (testnet only - Sepolia, Hoodi)

This product delivered verification at roughly 2,100 gas per proof at launch, a more than 10x improvement over Ethereum's native capacity with over 90% cost savings, and demonstrated that a decentralized network of operators could verify proofs at throughputs in the hundreds per second. ZK Arcade and other applications were built on top of this infrastructure during the testnet and mainnet phases, confirming that the architecture worked end-to-end.

Timeline

February 2024: The development begins.
June 2024: Public testnet launches as our first EigenLayer (now EigenCloud) AVS open to external operators.
November 2024: Mainnet beta goes live on Ethereum.
July 2026: Proof Verification Layer deprecation announced.

Instructions for operators

Over the past few days, we have been in direct contact with our operators to communicate this decision. In coordination with them, tasks are no longer being sent to our Proof Verification Layer, and they have been invited to deregister from it and shut down the instances currently running them. As part of this transition, many have begun deregistering, and our explorer will be updated to show only data from the Proof Aggregation Service going forward.

If you are currently running an operator on Aligned and have not yet done so, please deregister and shut down your instance as soon as possible. Thank you to the 59 operators who supported this product for the work you contributed to the network.

Conclusion

We are more convinced than ever that the future of the internet is going to be provable, and that ZK will help get us there. We said from the beginning that low-cost, high-throughput verification was needed for widespread ZK adoption and for Ethereum to scale, and that thesis has only gotten stronger. What has changed is where that verification should live and how it should be delivered.

The maturation of ZK proving systems, the trajectory of Ethereum toward native ZK support, and the shape of real-world demand all point in a consistent direction: the next chapter of this space is going to be built on top of stronger, more integrated infrastructure than existed when we started.

This is why our Proof Aggregation Service will become the default and only solution going forward.