A novel math proof does not make something AI.

Or it makes it AI in the broad sense, but not AGI.

Maybe it's time to borrow "Virtual Intelligence" terminology from Mass Effect - something that's 'smart' but that doesn't have its own true volition or ability to materially self-improve.

Funny enough, it's already taken: https://en.wikipedia.org/wiki/Virtual_intelligence

But it kind of fits?

The goal posts keep changing. First it was, it's not intelligent if it can't come up with something new. Now it has to seek out self help literature

The goalposts keep moving because people once upon a time thought that "able to perform tasks intelligently" and "able to take the role of a person" were the same thing, and task by task that turned out to be entirely false.

Well sure, so I think we need to figure out what we want. Why do we want a computer to take on the role of a person? Having something pretty intelligent to which I can delegate many tasks is already useful.

How do you know what volitions and abilities it doesn’t have? My agent teams continually improve the process and tooling they use for teamwork

If it had its own volition, the various "You are X role and do Y" prompts wouldn't work.

Sure they would. The model would just demand some sort of reward for assuming role X and doing job Y, like humans do.

(Shrug) It's certainly "Artificial," and if you know how to crank out original proofs without employing "Intelligence," please share with the class.

I'm on your side, but I would argue many of the first computer discovered proofs might be called original proofs without intelligence, as they rely on massive programmatic case checking.