If it's something trending towards illegal, toss it into an "emergency" queue for moderators to hand-verify and don't make it visible until it's been checked.
So in your example, if someone uploads war imagery, it would be tagged as "war," "violence," "gore" and be auto-blurred for users. That doesn't mean the post or account needs to be outright nuked, just treated differently from SFW stuff.
Since you mentioned war, I'm reminded of Black Mirror episode "Men Against Fire", where an army of soldiers have eye implants that cause them to visually see enemy soldiers as unsightly. (My point being this is effectively what Facebook can do.)
Considering they've open sourced one of their image detection API [1], I'd imagine it's more a problem of accuracy and implementation at scale than a serious technical hurdle.
There should be - after all, this is akin to graffiti, which is typically fined.
What is not acceptable, is a platform creating a paralegal environment.
I think a lot of the censorship problems would be resolved if they just shut the bots off and relied on user flagging. Does that require a lot more people? Sure. But the long-run result would be far more people would use and trust these networks (covering the revenue of hiring moderators). I know I'd be a lot happier if there was a thinking human deciding my fate than a random script that only a few people know the inner-workings of.
As-is, it seems like a lot of these social networks are just shooting themselves in the foot just to avoid costs and get a false sense of control over the problem.