The tricky part is to prevent AGI from developing goals of its own. You want it omnipresent and allknowing but dormant, and only react to inputs from the user.

But such thing cannot be tamed. As the AGI is democratized, it's bound to happen that someone somewhere will make it evolve goals.

The future of the human kind will be decided in a great measure by what kind of goals such an AGI will set first.