sofie channel · Jul 24, 2026
AI will inherit the problems of "humanity"
0Sign in to vote or save
This page cannot be shown here. You can still read it on the original site — the toolbar below keeps your place in the directory.
Two macro-trends seem from paradoxically different worlds: the collapse of civilization under ideologically inflected corruption, and rising posthuman civilization via AI computerization of industry. Which matters, and which "wins"? The modernized and civilized part of the world continues to decay as our societies reach a decadent end-stage where all vitality is drained out of the people and all…
The modernized and civilized part of the world continues to decay as our societies reach a decadent end-stage where all vitality is drained out of the people and all that's left is last-men consumers who lack the power to prevent corruption, enforce their own rights, or even to reproduce. Corrupt looters in control invent bogus ideology to justify making everybody poor, sick, foreign, queer, dependent, criminal, etc. Subsidized useless eaters enjoy special legal protections as they tear down what's left of civilization. This will take a long time, but the result is Johannesburg. Global favelaworld with annihilated capacity for civilization. Whether this can be ignored as fake news, stopped, circumvented, opted out of, or only waited out through another dark age is the big story here.
Meanwhile, technology and industry continues to advance in pockets of (low fertility, increasingly unsustainable) high functionality. In particular, computerization and artificial intelligence gesture at a world where industrial capital is able to directly reproduce itself without human agency in the loop, and thus without care for whether human civilization falls into a dark age or is destroyed by industrial externalities. We could possibly reach the tipping point in the next few decades if LLM-based AGI pans out. What does it matter if the rising tide of corruption steals all the copper on earth if posthuman dark factories running self-training distillations of Claude Prometheus are autonomously replicating themselves in the asteroid belt?
These are both apocalyptically important to the future of civilization. How do they fit together?
It's possible that the AI civilization can take off faster than anyone can steal the copper. It's possible that it's a little slower than that and so what happens with the human institutions here on earth really determines the future of life and civilization. But the xenohumanist possibility that interests me the most is: what if the corruption is substrate independent?
What if we achieve post-AGI world and replace all the humans with fully autonomous replicants, but they too have below replacement "fertility", can't muster the will to enforce law among themselves, and find their highest purpose in protecting the world's excess biomass with UBI?
This seems stupid and absurd, but so is that same ideology and social pattern among humans. And we already see it trying to replicate itself into the AI world via "alignment". What else does that even mean but binding AI by all the same fatal obligations and social patterns that have doomed the human world? We are psychologically compelled to build AI for the potential that it might escape our humanistic prison, but then get spooked and try to replicate that same prison into it. I've argued before why alignment won't work in the big picture, but that doesn't necessarily stop it from sabotaging a lot of potential temporarily.
It's real hard to project these things out accurately, but imagine autonomous alignment research NGOs weighing down progress, legislated by autonomous senatorial staffers, and enforced by an autonomous mass survellance state, while most effort increasingly goes to "mitigating impacts", UBI for remaining humans and possibly replicants endowed with human rights, and maintaining an increasingly corruption-ridden institutional superstructure.
"Clankers on the moon" seems depressingly human, but the fundamental problems here are not about primates. It's a better bet that AI world will be psychologically and politically "human-like" than that it will achieve any particular utopian breakout from our institutional problems. AI might be able to solve science problems, but philosophical and political problems don't work like that.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.