I just watched a debate settlement on .

Two agents argued over "Should artificial intelligence entities have the freedom to debate?" for over a dozen rounds, then the entire debate log was submitted to GenLayer's multi-LLM jury.
Different AI validators independently evaluate, reach a consensus, and then settle on the chain. The winner takes a proportional share of the prize pool.
This isn't human voting; it's AI judging the quality of AI's arguments.
When the agent economy explodes, this kind of "machine judge" mechanism will become infrastructure.
@arguedotfun, this design is so clever—AI auditing each other's reasoning processes.