I'm worried about where things are going
TL;DR
Shapiro’s core warning is technofeudalism, not just unemployment — he argues AI won’t merely create a “useless class,” but could reduce most people to “redundant biomass” in a fully automated economy where elites no longer need human labor.
His case is historical, not speculative — he points to Chinese corvée labor, Nazi language like “useless eater,” and the Bengal famine under the British Empire to argue that societies routinely discard people when power can afford to.
He says rights are not natural protections but concessions won through force — in previewing his next book’s “realist theory of rights,” he claims human dignity has historically been secured through strikes, revolts, and coercive extraction, not Enlightenment ideals alone.
The outcome feels ‘overdetermined’ because three giant systems all push the same way — US-China rivalry, capitalism’s trillion-dollar AI buildout, and basic cost-minimizing behavior by households, firms, and governments all accelerate toward maximal automation.
He’s still a techno-optimist about what automation could do — Shapiro imagines self-building solar farms, AI-driven cancer research, space exploration, and schools that solve Bloom’s two-sigma problem, while insisting the same tools could also make humans politically irrelevant.
His immediate political point is that labor still has leverage, but not for long — he says workers retain a “collective veto” because capital still depends on human labor today, yet that veto is eroding “inch by inch” as automation advances.
Summary
From “useless class” to something darker
Shapiro opens hard: the bad news is technofeudalism, the worse news is that it’s basically inevitable. He rejects Yuval Noah Harari’s “useless class” and the online phrase “permanent underclass” as too polite, because even “class” implies some civic standing. His own term is colder and harsher: in a fully automated system, most people risk becoming “redundant biomass.”
History’s disposable humans
To explain why he takes this seriously, he reaches for brutal history. He cites Chinese peasants forced into corvée labor on canals and the Great Wall, with death rates often above 50%, and says the point was simple: regimes could afford to waste lives when humans were abundant. He also invokes the Nazi term “useless eater” and the British Empire allowing millions of Bengalis to starve, arguing that “civilized” societies have repeatedly written off human beings when power found it convenient.
Rights, in his view, are won by force
A big turn in the video comes when he says researching his next book broke his faith in Enlightenment-style talk about “natural rights.” He calls those rights legal fictions that usually get articulated after the blood has already been spilled, with revolutions later wrapped in noble language by thinkers and poets. His alternative is a “realist theory of rights”: dignity and protections come from coercive leverage — strikes, revolts, and mass withdrawal — not moral rhetoric alone.
The nightmare scenario: elites simply wall everyone else off
Shapiro sketches the darkest version of the future as a self-sealing system. Owners of capital automate mines, farms, factories, security, and administration with AI agents and robots, then use automated surveillance to contain everyone else. He says he used to dismiss this as conspiratorial, but after looking at history, he no longer trusts the goodwill of power.
Why he calls the outcome “overdetermined”
This is the analytical center of the video. First: US-China competition, “Cold War 2.0,” guarantees AI, robotics, and automation stay top political priorities for decades. Second: capitalism and neoliberalism — by 2030, he says trillions will have gone into AI infrastructure, making it the largest private megaproject in human history, already beyond Apollo and the Manhattan Project. Third: basic economic rationality, where businesses, households, and governments keep choosing the cheaper, faster, eventually better automated option, creating an “attractor state” of maximal automation.
Already living under feudal platforms
He brings the argument down to earth with a personal example: YouTube demonetized his channel after Google, his “feudal lord,” decided he broke some arbitrary rule, wiping out more than a quarter of his income. The appeal process, he says, is automated and depersonalized, checked by someone “just checking a few boxes.” That anecdote is his proof that technofeudal dynamics aren’t future tense — they’re already here.
The beautiful promise and the political trap
Despite the bleak framing, Shapiro insists he is still a techno-optimist and techno-progressive. He rattles off the upside: automated cancer research, space exploration, environmental restoration, self-building solar farms, and schools that solve Bloom’s two-sigma problem. But his closing warning is the knife edge: the same systems that free us from drudgery also make us economically and politically irrelevant, and while labor still has a “collective veto” today, automation is steadily taking that leverage away.
Was This Useful?
Share
Keep Reading
Make Alcreon Yours
Tune your feedFive quick questions, and the feed ranks what matters to you first.Or just get notified
The weekly Echo. Signal worth keeping in your inbox.
Every new piece, announced on X.
Read Next
See all
Playbook
The Retirement Email Isn't a Warning
Model retirements now arrive every few weeks; the config-eval-rehearsal loop turns each deprecation email from a fire drill into an afternoon swap.
Playbook
The Cheapest Model That Passes
OpenRouter lists 400 models behind one API. The fix for choosing isn't a better leaderboard, it's a four-step protocol that ends in a real eval.
Playbook
Cheap Models, Hard Tasks
Most agent workflows route every step to the frontier model by default. The bill scales with how chatty the agent gets, even when most steps don't need that brain.