Fixed rules are the reason the AIs do well in games. You can’t train an algorithm to think outside the box like that, because the box is how it’s trained.
Since we cannot meaningfully codify, much less rapidly simulate, the potential actions and real consequences in this supposed global all-fronts war, we cannot train an algorithm on it.
Since we cannot meaningfully codify, much less rapidly simulate, the potential actions and real consequences in this supposed global all-fronts war, we cannot train an algorithm on it.