I once attended a talk a year ago where a techlead did just that - they had AI agents that ran a scrum team with different roles, each agent's prompt was to disagree with everyone else (or be highly critical) and present their own point of view, and then an arbiter would make the final decision. They claimed it worked for them.

Maybe. Humans form teams for a reason. Yes there are different exepriences and points of view in a human (vs. Not so much in LLM), but sometimes a different hat it all it takes. E.g. Code reviewer vs. Coder.

Seems likely to me. As long as adding more appears to help people will do it

Presumably there is some point where it levels out. And no doubt there will be a committee of AIs to determine said point.

Cause we wouldn’t want to boil the ocean…