CISA as the top 2026 priority is a bigger deal than most people realize. The fee savings from aggregating signatures across inputs aren't just a nice optimization. They fundamentally change the economics of collaborative transactions.

Right now a 2-input payjoin costs roughly the same in fees as two separate transactions because each input carries its own signature weight. With CISA, you'd collapse those signatures into one, saving around 7.5% on a typical 2-in-2-out transaction. That percentage grows with more inputs. A 10-input coinjoin batch could see 20-30% fee reduction.

The interesting second-order effect: cheaper payjoins mean more people actually use them, which means more transactions look identical on-chain. Privacy becomes the default because it's the cheap option, not the expensive one. That's the only way privacy tools ever get mass adoption. Nobody chooses the more expensive version out of principle for long.

Fabian's ASMap work is underappreciated too. Most node operators have no idea their peer selection is clustered by AS number. A single ISP going down (or being compelled) could partition a surprising chunk of the network. Embedding the AS map directly means every node gets geographic diversity in its peer connections without operators needing to think about it.

Good to see Brink funding long-horizon work like this. The CISA proof-of-concept with payjoin will be one of the most closely watched PRs of the year.