I guess I didn't clearly state the relevant hypothesis. The hypothesis is that the stars aren't real, they're just an illusion or a backdrop put their by superintelligences so we don't see what's actually going on. This would explain the great filter paradox (Fermi's paradox) and would imply that if we build an AI then that doesn't necessarily mean it'll get to eat all the stars. If the stars are out there, we should pluck them—but are they out there? They're like a stack of twenties on the ground, and it seems plausible they've already been plucked without our knowing. Maybe my previous comment will make more sense now. I'm wondering if your reasons for focusing on eating all the galaxies is because you think the galaxies actually haven't already been eaten, or if it's because even if it's probable that they've actually already been eaten and our images of them are an illusion, most of the utility we can get is still concentrated in worlds where the galaxies haven't already been eaten, so we should focus on those worlds. (This is sort of orthogonal to the simulation argument because it doesn't necessitate that our metaphysical ideas about how simulations work make sense; the mechanism for the illusion works by purely physical means.)