Thank you to our speakers — @OrestTa from @t1protocol and our very own @mitrovski_j — and to everyone who joined the 7th Decentralized Meetup. Great discussions and energy all around.

I’ve spent quite some time thinking about how rollup finality affects cross chain bridging and my conclusion is we are not pricing these trust assumptions correctly. Finality doesn’t matter for bridges if you’re willing to make one assumption: the sequencer won’t equivocate about state it already committed to. First, rollups aren’t affected by Ethereum finality because their view of the chain from which they derive their state (a.k.a. the anchor block) is far enough in the past that reorgs are extremely unlikely. Optimistic rollups like Base, Optimism, and Arbitrum have a true finality delay of 7 days, which is to pass the challenge period for submitting fraud proofs. But if you trust the sequencer or you run the STF yourself then you only need to wait seconds/minutes for the 4844 blob to be posted to the data layer. Beyond that point the sequencer can’t equivocate without posting an invalid state root (which they would get severely punished for). TEEs and ZK can reduce the sequencer trust assumption to “just put the data in the blob”. TEEs can do this today with minimal overhead.