Optimiums and the ‘L2’: Ethereum security revisited

The time period “Ethereum layer-2” is mostly thought to imply {that a} chain inherits Ethereum’s safety. However the line has at all times been blurry, and stays so, particularly with the rise of OP Stack chains that use different information availability (DA) layers. These so-called “Optimiums” settle to Ethereum however submit information elsewhere, introducing some further trust-assumptions.

Toghrul Maharramov, a rollup researcher and former Scroll contributor, has been a vocal critic of looser L2 definitions, and declared Wednesday on X:

“Optimiums usually are not L2s. Anybody who says in any other case is attempting to psyop you.”

An excellent instance of that is Celo, which Blockworks reported as a case examine of an L1 changing into an L2. For essentially the most half, Ethereum rollups submit their information to Ethereum itself — initially through calldata and now, predominantly, through blobs. That permits the helpful property of permissionless state reconstruction and withdrawals from an L2 to Ethereum mainnet.

Optimiums, in contrast, outsource DA to off-chain networks like EigenDA, Celestia or Avail, and depend on attestations to show that information is accessible. They could use fraud proofs and settle to Ethereum, however they add a layer of belief that Ethereum can’t immediately confirm.

Vitalik Buterin has used “L2” liberally, and certainly particularly referred to Celo as transferring to an “L2,” so we’re in good firm there, however the distinction is greater than semantics.

Ethereum scaling tracker L2BEAT now classifies most Optimiums as “Others,” a catch-all class for chains that don’t meet the strict rollup standards. Celo, for instance, accomplished its transition in March 2025 to run on the OP Stack with EigenDA.

In accordance with L2BEAT, it fails two key checks for rollup standing:

• It has fewer than 5 exterior actors who can submit fraud proofs
• It has no DA bridge that posts EigenDA availability proofs on Ethereum

The outcome: no L1-enforced treatment if information goes lacking.

Lisk, which migrated to Ethereum in late 2024, additionally falls into this bucket for related causes — reliance on a permissioned challenger set — though it does submit information on Ethereum.

In accordance with Sreeram Kannan, founding father of EigenLayer, we are able to consider rollups as “out-of-protocol scaling for execution,” whereas EigenDA is “out-of-protocol scaling for blob house.” In each circumstances, the info and logic occur exterior Ethereum’s protocol guidelines — however solely rollups guarantee Ethereum can implement validity immediately.

“When a certificates from EigenDA reveals up, you’re trusting that that certificates implies that the info is accessible,” he informed Blockworks.

That belief mannequin has no on-chain enforcement as we speak.

“If the DA claimed the info is accessible and the info just isn’t obtainable, there’s no recourse,” Kannan mentioned.

So, is “L2” simply Ethereum-aligned advertising? Or ought to it mirror the protocol’s potential to implement each correctness and availability?

L2BEAT’s present stance is that Ethereum should be capable to reject dangerous state roots and confirm information availability independently. The rest requires extra belief — whether or not that’s in a DA quorum, a governance council, or a permissioned set of challengers.

Optimiums with out both state validation or a DA bridge stay within the “Others” class. However because the tech matures, some could graduate to “Validiums & Optimiums” and even acquire rollup standing. To do this, you’d want some sturdy fallback mechanism or a strategy to make DA failures an “objectively attributable fault,” — a tricky drawback to crack.