I'd say Maslow's hierarchy[0] is a great starting point. Program that properly and faithfully (no backdoors, military exceptions, etc whatsoever) along with Asimov's 3 laws[1] and it should be pretty hard to find issue with the system that would result.
This is the "draw the rest of the owl"* of the alignment problem.
Or possibly the rest-of-owl of AI in general: Consider that there's still no level-5 self driving cars, despite road traffic law existing and the developers knowing about it since before they started trying.
The film version of I Robot had this right, the three laws are a manifesto for totalitarianism. The AI cannot sit on the sidelines as long as there is anything it can do to prevent crimes or abuse of any kind, no matter how intrusive that intervention may be.
This is the "draw the rest of the owl"* of the alignment problem.
Or possibly the rest-of-owl of AI in general: Consider that there's still no level-5 self driving cars, despite road traffic law existing and the developers knowing about it since before they started trying.
* https://knowyourmeme.com/photos/572078-how-to-draw-an-owl