"All goals"? Whose goals?
Not only do you need to prove that Decision Theory is inevitable, the same problem remains because "utility" is poorly defined even with 2 entities. This is a fundamental problem with Utilitarian ethics:
1. there are a lot of functions to pick and you need to prove the utility function any sufficiently advanced AI uses must itself be something we'd consider "nice"
2. utility monsters are a problem (starting at 2 entities)
3. the repugnant conclusion shows we don't know how to handle something as simple as adding more entities to the environment even when nobody's a utility monster
My point with historical monsters is that they can be competent without ever caring about the harms they cause.
You can be optimally efficient at your own goals without other minds in this universe sharing those goals. Pol Pot won't illustrate this because of how useless his regime was, but there's plenty of other monsters who would be a better fit here, e.g. Stalin. Or, from the perspective of turkeys, Bernard Matthews.
Any entity, E, who picks game-theoretic optimal decisions, will only care about the impact of their choices on others in their environment to the extent that the impact becomes more reward for E.
You need to show that an agent will inevitably care, despite the evidence of dangerously competent humans who don't, i.e. show that E must include the welfare of all who are not-E in their own utility.
You also need to show that whatever that utility function is, we agree with E's idea of our welfare.