Ethereum wants to check its own helper chains. The 15-minute wait is officially on notice.

Ethereum's L2 story is getting less modular. Native rollups and single-slot finality now point at the same bet. Ethereum wants to verify its own L2s, then settle that answer in seconds.

That's bigger than another scaling roadmap. The code path would move L2 trust away from custom verifier contracts and into Ethereum itself. The tradeoff is tighter L1 coupling for fewer custom security assumptions. If you're using an L2 because it feels like Ethereum, this is the part that decides whether that feeling becomes true.

EIP-8079, drafted in November 2025, gives the native-rollup idea a formal shape. The previous name was enshrined rollups, which was accurate but loaded. The mechanism is simpler to say than to ship: Ethereum would verify L2 execution directly.

Today, every serious rollup carries its own proof and verifier machinery. Bankless reports that Taiko's current setup uses 6 verifier contracts. Luca Donno of L2BEAT estimates native rollups could cut onchain verifier code by about 39%. That number is single-sourced, so treat it as directional, not gospel.

The direction still matters. Less verifier code means fewer places where an L2 can be Ethereum-like in marketing and app-specific in security. At the contract level, this turns a rollup from a bridge and verifier setup into something closer to an Ethereum execution lane.

The ethrex client team has already shown a working demo. Bankless describes L2-to-L1 settlement through re-execution, including deposits and withdrawals. That is the right demo to watch. Token transfers are where architecture stops being a diagram.

Fast finality is the second half. Ethereum blocks arrive about every 12 seconds, according to Bankless. Full finality still takes about 15 minutes. That delay is invisible until your bridge, exchange, or app decides it isn't.

Francesco D'Amato's fast-confirmation rule targets that gap. Bankless says the rule is live on Glamsterdam devnets. In a replay of one year of mainnet data, it reached 1-slot confirmation more than 95% of the time. The same report says it produced 0 false confirmations.

Those numbers are not finality. They are a rule for when a block looks safe enough early. The difference matters. This is a cache invalidation problem with money attached: acting early is useful only if your invalidation rate stays near zero.

Put the two arcs together and the shape is clear. Native rollups reduce the custom verifier surface. Fast confirmations reduce the waiting window. The result is not just faster L2s. It is Ethereum trying to make the L1 and L2 split feel like blockspace allocation, not separate trust domains.

The engineers know this; the marketing team pretends they don't. Rollups won because Ethereum did not have enough cheap blockspace. They also created a weird social contract where users treat many L2s as Ethereum, while the code says each stack has its own moving parts.

Native rollups challenge that bargain. They ask whether Ethereum should absorb the verification role instead of outsourcing it to every rollup team. That is cleaner for users, but harder for protocol design. Ethereum would be taking more responsibility for what used to live at the edge.

Barnabé Monnot's end-state framing goes even further. The L1 could eventually verify its own blocks through proofs, making Ethereum something like a rollup of itself. There is no need to over-mystify that. It means the same proof logic that checks L2 execution could also tighten L1 execution.

The timeline is still long. L2BEAT has opened a Native Rollups tracker with milestones through 2027. Bankless reports a devnet target for December 2026. So the only number that matters today is not 39% or 95%. It is whether real settlement works outside a demo.

Read the PR diff, not the thread. If native rollups reach public devnets with deposits, withdrawals, and re-execution intact, custom verifier stacks start looking temporary. If fast confirmations keep showing zero false confirmations under real devnet pressure, the 15-minute wait starts looking temporary too.

That is the actual convergence. Ethereum is not killing L2s. It is trying to make them boring. For infrastructure, boring is usually the highest compliment.