It seems like it depends on what kind of AI is put in charge by who and how ideally there would be literally only upsides.

It seems like it depends on what kind of AI is put in charge by who and how. Ideally there would be literally only upsides.

Ai is a tool. Governemts use tools just as everybody else. And they pay more for their tools.

Reg tech is a thing.

True, would it be better to utilize this tool as much as possible? Essentially automating every aspect of government?

So essentially as long as ethics and oversight are strongly influenced in implementing AI, we could completely automate government? Or should there be some level of human input?

Depends on the type of AI in my opinion. Complete removal of the human element can only really be done with an ai comparable to god in my opinion, with everything else you will need some level of collective human control in my opinion. It’s my belief that the correct way to solve many of todays problems is just to advance so fast that and far that things that would be difficult today are trivial, at least until aforementioned god AI comes along.