suppose that a Friendly AI fills a human-sized three-dimensional grid with atoms, using a quantum dice to determine which atom occupies each "pixel" in the grid. This splits the universe into as many branches as there are possible permutations of the grid (presumably a lot)

How is that a Friendly AI?

'alive' relative to a specific environment

It's always relative to a certain environment. Human beings and most animals can't survive outside of the current biosphere. In that respect we're no less independent from certain peculiar conditions than viruses are. We both depend on other living organisms in order to survive.

Maybe redefine life against a continuum of how unlikely, complex the necessary environmental conditions are that sustain it?

Some autotrophic cells might rank at one currently known bound while higher animals would be on the other end.

A faith which cannot survive collision with the truth is not worth many regrets.

Arthur C. Clarke

Yeah being considered a part of an AI. I might hate to be,say, its "hair". Just thinking about its next metaphorical "fashion induced haircut and coloring" gives me the chills.

Just because something is a part of something else doesn't mean it'll be treated in ways that it finds acceptable, let alone pleasant.

The idea may be interesting for human-like minds and ems derived from humans - and even then still dangerous. I don't see how that could apply in any marginally useful way to minds in general.

For what it's worth I had already observed this effect. I am less likely to carry on with some plan if I talk about it to other people. Now I tend to just do what I have to, and only talk about it once it's done.

Part of the problem is I hate feeling pressured into doing something. Social commitment will, if anything, simply make me want to run away from what I just implicitly promised I'd do. Perhaps because I can never be sure whether I can achieve something : if I fail silently and nobody knows, it's ok. Less so if I told people about it. It feels better to run away from something (failing by choice) than failing for other reasons.

Also in some cases, just saying you plan to do something already feels like you've done something. Either because you count it as a step towards doing the whole thing (a step after which it feels more acceptable to take a break, which can last indefinitely long), either because you fantasized about it enough that you don't feel the need to implement it for real anymore.

I feel like I can relate to that. It's not like I never rationalize, but I always know when I do it. Sometimes It may be pretty faint, but I'll still be aware of it. Whether I allow myself to proceed with justifying a false belief depends on the context. Sometimes it just feels uncomfortable enough to admit to being wrong, sometimes it is efficient to mislead people, and so on.