for small groups nearly every dispute could pose existential risk, but for enormous groups very few disputes pose existential risk. then in the conflict side, enormous groups have MAD, where conflict itself carries existential risk. small groups of course can also be wiped out by conflict (even if they don't possess a bureaucratic mechanism to ensure that,) but the risk side of the calculation is heavier for them I think.
and enormous groups still engage in conflict with each other although it happens indirectly via proxy wars.
also paradoxically the larger the group becomes the more its interests, or "meta interests" (such as favoring stability and security), tend to align with its competitors but for small groups it's almost like every group for itself. but I believe that the calculation can tend towards cooperation being more beneficial the larger group gets. but I think this factor only applies when groups become very very large.
also there's some sort of entropy argument as in large groups cultivate more structure, organization and order and therefore it would take more energy to dismantle them so they're more stable. which not only makes them harder to dismantle but also gives them more to lose if they were to decide to tear everything down and build it up again.
that's how I think about it. what do you think?