TechFlow Logo
Login/ Sign up
ETH Gas
Gwei
Fear
gas
Post-Quantum Timeline Delayed by Ethereum: Is Starknet Trying to Set Its Own Pace?

Post-Quantum Timeline Delayed by Ethereum: Is Starknet Trying to Set Its Own Pace?

2026.10.15
Share
A quantum-resistant network by 2027? Starknet's standalone L1 bet.

Written by: Bankless

Compiled by: AididiaoJP, Foresight News

Could mathematical breakthroughs in quantum computing and AI cause public chain cryptography to become obsolete prematurely? This question has been reignited by Justin Drake's comments on "bunker mode." Starknet subsequently presented a more concrete move: considering transitioning from an Ethereum L2 to an independent L1, with the goal of becoming the first fully post-quantum network by 2027.

The reasoning is straightforward: as an L2, its post-quantum upgrade pace is bottlenecked by Ethereum.

Under the Dual Shadow of Quantum Computing and AI, Drake Calls for "Bunker Mode"

On October 7, Justin Drake urged the blockchain industry to calmly initiate planning for "bunker mode." His personal recommendation is to orderly drive a large-scale migration of assets to new addresses, keeping public keys hidden behind hashes, starting with large holders.

Whether these judgments underestimate or overestimate the timeline for mathematical breakthroughs can only be proven by time. But both quantum and AI threats are real, not distant or abstract risks. Some crypto projects are clearly taking this more seriously than others.

Starknet is one of them. Last week, its developers announced they are considering migrating the chain from an Ethereum L2 to an independent L1, sparking attention. On October 8, Starknet's official account tweeted: "We are actively considering becoming an L1." The official statement noted that this would allow Starknet to become the first fully post-quantum network, targeting 2027.

Later that day, StarkWare CEO Eli Ben-Sasson posed a question: Is it a good idea or a bad idea for Starknet to become an L1 in exchange for post-quantum agility? He highlighted two things that must be watched: the quantum threat may be much closer than expected; and AI is advancing faster than we can handle.

On October 11, he further explained: Q-Day is already a severe threat, and now with AI added, the timeline will shift. The logic is that if Starknet becomes an L1, it can dictate its own pace for security upgrades rather than following Ethereum's lead.

Where Starknet's Post-Quantum Timeline Stalls as an L2

Why does staying on an L2 prevent achieving this goal? The roadmap published on June 30 explains why.

Starknet's quantum migration is divided into three phases. The first two phases are executed solely by Starknet: replacing Pedersen hashes with post-quantum BLAKE2, integrating post-quantum signatures like Falcon-512, and migrating legacy contracts. The third phase falls outside its control because it involves two components borrowed from Ethereum.

The first is bridge messages. These messages use elliptic curve calls that are vulnerable to quantum attacks when transmitted between chains. StarkWare stated plainly: removing these calls before Ethereum's migration would break the cross-chain bridge.

The second is Blob data. Starknet publishes state data to Ethereum via Blobs, which are anchored by elliptic curve-based KZG commitments. If KZG is compromised, according to StarkWare, it affects data availability, not the reliability of Starknet's state itself—which remains protected by STARK proofs. However, the June roadmap leaves this fix entirely to Ethereum.

Ethereum may still need several years to complete its own post-quantum evolution. This means at least one critical component in Starknet's defense roadmap will be frozen for that long.

This is not a problem unique to Starknet. Rollups inherit the L1's cryptography and its security model, meaning their quantum timelines effectively depend on whichever is slower: themselves or the L1.

Starknet anticipates hitting this wall because its half of the work could proceed much faster.

Ethereum's Quantum Roadmap Is Ironically Becoming More Like Starknet

More interestingly, Ethereum's own quantum plans are making it increasingly resemble Starknet.

For example, Ethereum is advancing EIP-8141, which includes allowing accounts to adopt quantum-safe signatures on their own schedule without waiting for a network-wide migration. Meanwhile, Starknet has implemented native account abstraction for years—every account is a programmable smart contract. Therefore, S2morrow can already build post-quantum Starknet accounts around Falcon-512 signatures without requiring protocol changes.

Ethereum also plans to replace BLS validator signatures with hash-based leanXMSS, and use leanVM to compress bulkier results via zk proofs. Among its listed alternatives to KZG, STARK-based commitments are among the leading candidates. STARKs are inherently hash-based, which is exactly what the "Stark" in Starknet's name stands for.

Over the past few years, Starknet has already pieced together many components needed to become an L1: STRK staking, multiple provers, and so on. Originally intended to make Starknet a better decentralized scaling solution for Ethereum, they have indeed significantly reduced the "L1 gap."

Currently, the gaps mainly consist of three areas: data availability, i.e., where transaction data is published so anyone can reconstruct the chain; final settlement; and customizing the official bridge as needed.

The Most Critical Open Question: Should It Still Reconnect to Ethereum?

When it comes to bridges, consider how today's StarkGate withdrawals work: funds are only released once Ethereum validates the proof for the withdrawal block and the corresponding state update has landed on Ethereum. In other words, Ethereum verifies Starknet's mathematical proofs, preventing Starknet operators from forging the state upon which bridged funds rely.

Once Starknet becomes an L1, it can position itself anywhere along the trust model spectrum from "dependent on Ethereum verification" to "fully independent": either continuing to submit STARK proofs to Ethereum to support a minimal-trust bridge, or going completely standalone. These two choices correspond to entirely different trust models, representing the key open question I will focus on.

Remember, Starknet has consistently stated it can settle on both Bitcoin and Ethereum in the future. But this does not mean leaving Ethereum. If it becomes an L1 and connects and bridges both chains through proofs, this actually aligns better with its original vision rather than deviating from it.

Not Yet Official, But the Direction Is Clear

No formal transition has been initiated yet, but an L1 architecture appears to be Starknet's path forward. Regardless, the chain wants to act early, because for projects that fail to adjust in time, there may be no "later."

This assessment holds water.