My simulation gambit paper identifies the Nash Equilibrium for why Effective Altruism is the default position for sufficiently intelligent agents: https://darayat.substack.com/p/why-asis-might-self-align-a-gambit