Pinned Tweet
Every privacy protocol says "private."
But the question is: private from whom?
We just shipped a new feature on L2BEAT that answers that for every protocol we track, against 5 different adversaries.
October was busy at L2BEAT.
We shipped three new pages, added more projects and metrics to Privacy, published research, and ran a couple of community initiatives.
A quick recap of what changed π
On the community side, @kaereste moderated a panel at @stblcnsummit .
The discussion covered how enterprises choose between stablecoins and tokenised deposits, and what changes when AI agents start initiating payments.
xitter.gq/stblcnsummit/status/21β¦
How are enterprises actually choosing between stablecoins and tokenised deposits for commercial settlement, and why does the viable instrument vary across markets and regulatory regimes?
What does enterprise settlement at volume actually require across accounting systems, reconciliation, and approvals, and what changes once an AI agent initiates the payment?
On stage:
Tushar Gulhane, @SAP
Mai Kaneko, DCP Co., Ltd.
Sanchit Mall, @Visa
Naojiro Hisada, Rakuten Group, Inc.
Moderator:
Krzysztof Urbanski, L2BEAT
8 October Β· sg26.stablecoinsummit.com
We've reviewed 16 privacy protocols to understand what they actually hide, who they hide it from, and what's still visible.
How do we assess them? We've published our full methodology, and we'd love your feedback π
But a good privacy rating doesn't mean much if you're the only one using a pool, for example.
That's why we also look at anonymity sets: how many unique users have deposited into a pool over the last 30 days. Some pools have 1,000+ depositors, others have barely any.
We also track trusted setup, exit windows, and reproducibility.
This is our approach so far, but it's not set in stone. Think we've missed something or should assess a threat differently?
Join the discussion on the forum π
forum.l2beat.com/t/privacy-aβ¦
What if @ethereum could verify ZK proofs itself, so rollups and apps wouldn't need to run their own verifiers?
This could be achieved through native proof verification.
And we've made some progress there over the past month: the draft for EIP-8357 was merged, and a native rollup is now running on a devnet π
There's still work to do before native rollups can run on Ethereum.
L1 needs leanVM, frame transactions, EIP-8288 and EIP-8357 included in a fork.
But with the first pieces coming together and a working devnet to experiment with, we're getting closer.
Want to dig deeper? We've put everything in one place: l2beat.com/native-rollups/





