The Tree of Arguments in a Debate?

There's a worry that smart systems could come up with strategies that humans erroneously find extremely convincing. This is still a concern, but i think we can do a lot better than what this looks like if you imagine just two agents having a free, a sort of free text debate where they say things back and forth. And also, there are a bunch of things you can do with m el sistens that youu can't do with humans that are pretty useful. Its like you can save a copy of a debater from earlier in the debate, when they don't know a what the later context is, and you can ask questions to them. It inforces having a consistent