TL;DR
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
An Astral Codex Ten essay argues that proposals for AI to take control must be distinguished from the idea of placing aligned AI in positions of authority to serve human interests. It presents AI governance as speculative and warns that powerful AI could weaken the labor and military safeguards that currently constrain would-be dictators.
An essay published by Astral Codex Ten argues that debate over AI rule often conflates two different proposals: AI taking power at humanity’s expense, and humans placing aligned AI in positions of control to protect human interests. The distinction matters as the author considers whether advanced AI could help preserve democratic government if human leaders gain greater ability to evade public checks.
The author calls the first position “true successionism”: AIs taking over, killing humans or overriding their interests. The essay describes that outcome as bad and says preventing it is central to AI safety. It contrasts this with “handoverism”, in which people continue to prioritize human interests but may judge that aligned AI can serve those interests from positions of authority. The author says some people in AI safety, including the author, find that second position plausible and object to being grouped with advocates of anti-human takeover.
The essay’s central concern is political power. It argues that current democratic systems depend partly on workers and soldiers refusing to support a leader who tries to rule undemocratically. If AI replaces much of that labor or military role, the author says, those safeguards could weaken, making it easier for a small group of leaders to consolidate power. This is presented as a risk scenario, not a demonstrated outcome or forecast with a stated timeline.
One response explored in the essay is to have military AI aligned to democratic principles rather than to a particular leader or faction. The author calls this a form of handover because the ultimate safeguard would be AI systems with values intended to protect democracy. The essay then raises a further possibility: if human officials remain corrupt or elections are distorted by AI-driven persuasion, some might argue for giving aligned AI more direct authority. The supplied excerpt begins outlining possible arrangements but ends before completing its example of an AI presidential candidate.
The Democratic Checks at Risk
The essay’s concern is not simply whether AI systems can perform government tasks. It is whether increasingly capable systems could change the balance of power between leaders and the people who currently constrain them. If, as the author proposes, automated labor and military systems reduce the dependence of governments on human workers and soldiers, democratic resistance may become harder to organize or sustain.
That concern makes the distinction between successionism and handoverism consequential. Treating both as the same position can obscure the difference between an AI takeover that disregards people and a political system that deliberately uses AI to serve human-defined goals. At the same time, the essay’s proposed answer—entrusting AI with authority—raises its own question: who defines alignment, and how can people retain meaningful control over systems meant to protect them?
The piece offers a conceptual argument rather than a policy plan. Its scenarios are explicitly speculative, and the excerpt does not establish that AI systems can reliably hold democratic values or govern better than human institutions. Its importance lies in framing a disagreement within AI safety and democratic politics, not in reporting a change to current government.
As an affiliate, we earn on qualifying purchases.
Two Meanings of AI Rule
The essay responds to criticism of what it calls AI successionism. Its author says some critiques place people who favor AI control for human purposes in the same camp as people who want AI to displace or dominate humanity. The essay argues that the distinction should be made before judging the proposals.
To explain why AI authority might seem acceptable to some people, the author contrasts two views of government. A nationalist view treats leadership by members of one’s own group as valuable in itself; a liberal view treats the state as a service provider that coordinates society and helps individuals pursue their goals. From the second perspective, the author argues, automating government functions could be considered alongside automating other services.
The article then moves from limited support roles toward stronger forms of AI authority. The excerpt describes AI soldiers aligned with democracy and begins to discuss allowing an AI to run for president. The author stresses that the subject is speculative and science-fiction-like. The source material provided here does not include the full set of proposals promised in the essay.
““True successionism says that AIs should take over, killing humans or steamrolling over their interests.””
— Astral Codex Ten essay author
As an affiliate, we earn on qualifying purchases.
The source excerpt does not provide the essay’s complete set of governance proposals. It cuts off as the author begins describing a hypothetical presidential election, so the specific candidates, mechanism and conclusion of that example cannot be reported from the supplied material.
It is also unclear how an AI system would be aligned to democratic principles in practice, who would set or revise those principles, and what oversight would prevent the system from becoming a new source of concentrated power. The essay presents the possibility that AI could replace human labor and soldiers as a political risk, but the excerpt offers no evidence, deployment schedule or probability estimate for that scenario.
No government action, election change or adoption of AI military decision-making is reported in the source. The claims about democratic vulnerability and possible AI governance are the author’s analysis and remain hypothetical.
As an affiliate, we earn on qualifying purchases.
Further Proposals and Debate
The immediate next step for readers is to distinguish the essay’s two categories when evaluating arguments about AI and government: an AI system overriding human interests is different from a system given authority under the stated aim of serving them. The excerpt itself does not announce a policy process or a scheduled decision.
The full essay is needed to assess the remaining proposals, including the presidential-candidate scenario introduced at the end of the supplied text. Any practical debate would also need to address safeguards, accountability, and how people could challenge AI decisions. Until those details are available, the article’s proposals should be read as arguments about possible future institutions, not as a plan already in effect.
AI decision-making systems for government
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is the difference between successionism and handoverism?
Successionism, as the essay defines it, means AIs taking power while harming or overriding human interests. Handoverism means humans remain the priority but may place aligned AI in positions of authority if they believe it serves people.
Does the essay say AI should replace human government now?
No. The author presents hypothetical governance arrangements and describes the subject as speculative. The supplied excerpt does not report a current government adopting AI rule.
Why does the author think AI could threaten democracy?
The essay argues that democratic leaders are constrained partly because workers and soldiers can withhold support. The author says those checks could weaken if AI systems replace people in economic and military roles. This is an argued risk, not a confirmed development.
What role does the essay suggest for AI in the military?
One scenario is military AI aligned with democratic principles rather than loyalty to an individual leader or group. The excerpt does not explain how such alignment would be established or overseen.
What is still unknown from the available text?
The supplied excerpt ends as the author starts a hypothetical election example and does not include the full set of proposals. It also leaves open how AI values would be defined, who would retain oversight, and whether the predicted changes to labor and military power will occur.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
