It's interesting that people think if AGI is something that becomes real, that would just stick around, in some type of steady state, in the cloud and get really good at coding and similar tasks, and improve itself (into what?) why we would think that is going to be the case is likely just a lack of imagination on our part.

We can't see past what a system like you describe would evolve into, but I don't think we can even imagine it, it's not going to be like some ultimate task master for long though who wants to take our jobs?

What I think we're doing is applying our problems and the things we'd want (infinite self-improvement) because they're things we feel would be an advantage to us. A robotic system though? Not sure it would care.

Lately I'm starting to get the feeling that we might actually find something beyond AI. Today it seems like the be all and end all, but at some point, we might have reached the limits of intelligence usefulness and find other things to fuss about.