a16z's Aerie challenges Ethereum's hash-based quantum plan with lattice signatures
a16z crypto researchers say Aerie aggregated 32,768 Falcon signatures in 0.65 seconds on a laptop, 29 times faster than leanVM on consensus signatures. Ethereum, which targets full quantum resistance by December 2029, has not committed to adopting it.
The Investor · Invest desk
Key figures from the researchers' benchmarks: Aerie was roughly 60 times faster than leanVM on stateless transaction signatures and 29 times faster on consensus signatures, both figures disputed, and aggregated more than 50,000 signatures per second on an Apple M4 Max laptop without GPU.
| Figure | Value | Note | Claim |
|---|---|---|---|
| Stateless transaction signatures vs leanVM, roughly | 60 times faster | disputed | 21 |
| Consensus signatures | 29 times faster | disputed | 20 |
| Signatures aggregated per second, more than | 50,000 signatures | Apple M4 Max laptop, no GPU | 2 |
What happened
- Aerie pairs lattice-based signatures with lattice-based proofs and aggregates thousands of signatures directly, without the recursive rounds Thaler objects to.
- Proof verification took 54 milliseconds on a single CPU core, and the output averaged about 42 bytes per signature.
- Ethereum relies on BLS signatures today, which combine thousands of validator approvals efficiently but are not considered resistant to sufficiently powerful quantum computers.
Why it matters
- Who has to decide decision Roadmap planners have about 38 months to settle whether the signature layer stays on recursive hash proofs or moves to lattices.
- What it limits constraint If Thaler's encryption point holds, a hash-only stack cannot deliver the roadmap's private-transactions goal, so some lattice component is likely whichever signature design wins.
- Who's exposed exposure Anyone planning around the 29-times and 60-times figures is relying on the authors' own laptop benchmarks, not results from a live network.
Next story · Invest · 4d ago
Hoskinson says Buterin's AI warning could pull developers off lattice cryptographyConfirmed2 publishers
The speed claim is easier to size than to trust. Aggregating 32,768 Falcon signatures in 0.65 seconds on a laptop with no GPU [2] comes to about 20 microseconds a signature [14]. The lead over leanVM is 29 times on consensus signatures in what the report calls comparable benchmarks [20] and roughly 60 times on stateless transaction signatures [21], so it about doubles between the two workloads [22]. Crypto Briefing notes that the figures come from the researchers' own published benchmarks and do not show how Aerie would perform on a live blockchain network [3].
The design disagreement is narrow. Ethereum researchers are building leanVM to compress signatures with cryptographic proofs and lighten the load on validators [10]. Thaler argues that the hash-based route requires multiple rounds of recursive proof aggregation and that this adds computational cost and complexity [8].
The stronger argument has nothing to do with speed. Established cryptographic research shows that, in a general black box setting, hash functions by themselves cannot be used to build public key encryption [16], and private transfers need encryption that works even when sender and recipient have never communicated [15]. The Ethereum Foundation's draft roadmap lists native private transactions beside quantum resistance as long-term priorities [12]. Thaler concludes that Ethereum may eventually need lattices or similar structured systems whether or not it uses them for signatures and proofs [17].
That last clause separates the two cases. We think the likeliest compromise is hash-based signatures through leanVM with lattices confined to encryption, and that view is wrong if independent benchmarks on a live network reproduce the 29-times lead, because a signature layer that slow would then be hard to defend. The counter-case is Thaler's own complexity objection, which we think would apply with equal force to a stack that runs two cryptographic families.
Lattices have costs of their own. The Aerie researchers acknowledge that Falcon signing has historically required complex arithmetic and substantial memory, though recent improvements have reduced those requirements [4]. Thaler also pushed back on fears that future AI advances might compromise lattice cryptography, saying any speculative attack has to be weighed against how complex the existing alternatives are [18]. Crypto Briefing does not report a response from the Ethereum Foundation or the leanVM developers.
Ethereum's proposed quantum-resistant architecture remains under development [13]. Until the Foundation says otherwise, a planner working to the December 2029 date is working to the hash-based design, with about 38 months left [19].
What to watch
- Any response from the Ethereum Foundation or the leanVM developers to Thaler's recursion and privacy arguments.
- Independent benchmarks of Aerie against leanVM, especially on a live network rather than one laptop.
- Whether the Foundation's draft roadmap changes how it plans encryption for native private transactions.
Information, not investment advice. This story is general information about companies and markets, drawn from cited reporting. It is not a recommendation to buy, sell or hold any security; check the cited sources before acting.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence55
- Adoption3
- Hype gap+25
- Incentives60
- Confidence70
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
Justin Thaler and a team of cryptographers at a16z crypto introduced Aerie in a Friday research announcement, a system designed to aggregate thousands of quantum resistant digital signatures into a single cryptographic proof.
- [2]
In tests on an Apple M4 Max laptop without GPU acceleration, Aerie aggregated 32,768 Falcon signatures in 0.65 seconds, equivalent to more than 50,000 signatures per second.
- [3]
The results are based on the researchers' published benchmarks and do not establish how the system would perform across a live blockchain network.
- [4]
The Aerie researchers acknowledge that Falcon signing has historically required complex arithmetic and substantial memory, although recent improvements have reduced those requirements.
- [5]
Ethereum has not committed to adopting Aerie.
- [6]
The resulting data averaged approximately 42 bytes per signature, while proof verification took 54 milliseconds on a single CPU core.
- [7]
Aerie combines lattice based signatures with lattice based cryptographic proofs, allowing thousands of signatures to be aggregated directly.
- [8]
Thaler argues that the hash based approach requires multiple rounds of recursive proof aggregation, adding computational costs and complexity.
- [9]
Ethereum currently relies on BLS signatures, which allow thousands of validator approvals to be combined efficiently, but the system is not considered resistant to sufficiently powerful quantum computers.
- [10]
Ethereum researchers have been developing leanVM, a system that uses cryptographic proofs to compress signatures and reduce the burden on validators.
- [11]
Ethereum is working toward making its consensus, execution and data layers resistant to potential quantum computing attacks by December 2029.
- [12]
The Ethereum Foundation's draft roadmap identifies both quantum resistant security and native private transactions as long term priorities.
- [13]
Ethereum's proposed quantum resistant architecture remains under development.
- [14]
32,768 signatures aggregated in 0.65 seconds is about 20 microseconds per signature.
- [15]
Private transfers require encryption so recipients can securely receive information about transactions, including when the sender and recipient have never communicated before; Thaler argues that relying exclusively on hash based cryptography cannot provide all the functionality required for private payments.
- [16]
Public key encryption cannot be constructed from hash functions alone in a general black box setting, according to established cryptographic research.
- [17]
Thaler argues that Ethereum may eventually need structured cryptographic systems, such as lattices, regardless of whether it adopts them for signatures and proofs.
- [18]
Thaler challenged concerns that future AI advances could compromise lattice cryptography, arguing that speculative attacks should be weighed against the complexity of existing alternatives.
- [19]
From October 2026 to December 2029 is about 38 months.
- [20]
According to the researchers, Aerie processed consensus signatures 29 times faster than Ethereum's leanVM in comparable benchmarks.
- [21]
The advantage over leanVM increased to roughly 60 times for stateless transaction signatures.
- [22]
The reported speed lead roughly doubles between consensus signatures (29 times) and stateless transaction signatures (roughly 60 times).
Sources
2 independent publishers whose own reporting we read for this story.
- crypto.newsLayerZero unveils Aerie with 0.65-second quantum-safe signature benchmark
1 article · October 16, 2026
- cryptobriefing.comLayerZero’s Aerie tackles the post-quantum signature bloat problem
2 articles · October 16, 2026
Topics and entities
Topics
- Signature aggregation
- Ethereum quantum-resistance roadmap
- Lattice-based cryptography
- Post-Quantum Cryptography