Actual CDSs are quite a bit more complicated than that. They pay out in the case of very specifically defined credit events, and it's not uncommonly a difficul

Actual CDSs are quite a bit more complicated than that. They pay out in the case of very specifically defined credit events, and it's not uncommonly a difficult question to decide whether a credit event has occurred. The current system requires 80% of the ISDA's Credit Derivatives Determinations Committees to vote that an event has occurred. You see to think a purely mechanical system would be better, but it's not obvious to me why this would be so. It would certainly be different. > Look ma, no more humans! Yes, but why is that good? Is the role of humans in the CDS process a problem? Why? A CDS is a contract between two parties; do you believe more automation in the process would benefit the seller, the buyer, both, or society at large? And how? The main advantage of the blockchain would seem to be its distributed nature and it's ability to not need any central trusted authority. But in the cases of CDSs, we have a central trusted authority in the form of the ISDA. Again, we come back to the issue of needing to begin by defining the problem we're trying to solve, and then demonstrate how a blockchain could solve it. What problem with the existing system are you trying to solve?