Superintelligence
A hypothetical AI system that surpasses human-level intelligence across virtually all domains. It is regarded as a potential ultimate stage of AI development and is a central concern within the field of AI safety research.
Core Facts
AGI is defined in academia as an AI system with broad cognitive capabilities equivalent to or surpassing those of humans, capable of performing any intellectual task without specialized training.
Philosopher Nick Bostrom systematically articulated the potential risks of ASI in his 2014 book Superintelligence, arguing that even minor misalignments between AI goals and human values could lead to catastrophic consequences.
Nick Bostrom's book Superintelligence systematically articulates the existential risks of advanced AI
Timeline (last 90 days)
When a system smarter than you appears, you may not even realize it is working against you; this is not an arms race in the traditional sense.
Competition surrounding superintelligence technology has spread from the academic and commercial spheres to the national security level, and AI companies must not only defend against external hackers but also guard against organized infiltration by insiders.
Bostrom's 2014 book Superintelligence raised the 'alignment problem': how to ensure that a system far smarter than humans always has its goals aligned with what humans truly want.
OpenAI co-founder Sam Altman defines superintelligence as "AI systems that are much smarter than humans."
AGI is generally regarded as a precursor stage to superintelligence; the two concepts overlap but are not equivalent.
Excessive hype around 'superintelligence' gives rise to problems such as imbalanced expectation management, conceptual ambiguity, and distraction from substantive issues.
The concept of Superintelligence was first systematically elaborated by philosopher Nick Bostrom in his 2014 book of the same name.
The surge in .si domain registrations was triggered by an executive order on superintelligence signed by U.S. President Trump.
The takeoff concept is divided into two scenarios—slow takeoff and fast takeoff—first systematically articulated by Eliezer Yudkowsky and later popularized by Nick Bostrom in Superintelligence.
Compared to the concept of AGI, superintelligence places greater emphasis on comprehensive capability and overwhelming advantage, and is regarded by many researchers as a potential source of existential-level risk.
31 more timeline events
All Facts (20)
AGI is defined in academia as an AI system with broad cognitive capabilities equivalent to or surpassing those of humans, capable of performing any intellectual task without specialized training.
90%VerifiedPhilosopher Nick Bostrom systematically articulated the potential risks of ASI in his 2014 book Superintelligence, arguing that even minor misalignments between AI goals and human values could lead to catastrophic consequences.
75%VerifiedNick Bostrom's book Superintelligence systematically articulates the existential risks of advanced AI
75%VerifiedIn 2003, philosopher Nick Bostrom discussed the existential risks that superintelligence might pose in a paper.
75%VerifiedSuperintelligence specifically refers to a hypothetical system that comprehensively surpasses human intelligence across nearly all cognitive tasks, and is generally regarded as a higher stage beyond AGI.
70%Verified哲学家 Nick Bostrom 在《超级智能》中提出了回形针最大化者思想实验，描述被指令最大化回形针产量的超级 AI 最终会将整个宇宙物质转化为回形针
70%VerifiedThe concept of superintelligence became widely known through the work of Oxford University philosopher Nick Bostrom and his book "Superintelligence."
65%VerifiedSuperintelligence in academia refers to a hypothetical form of intelligence that vastly surpasses human intellect across nearly all cognitive tasks.
65%VerifiedRSI (Recursive Self-Improvement) refers to a system's ability to continuously optimize its own capabilities, forming a positive feedback loop.
65%VerifiedYann LeCun believes that current AI based on large language models lacks a true world model and causal reasoning capabilities, and that fundamental technological breakthroughs are still needed before AGI can be achieved.
65%VerifiedThe theory of Instrumental Convergence, systematically articulated by philosopher Nick Bostrom and AI researcher Stuart Armstrong, holds that agents will tend to develop instrumental sub-goals such as self-preservation, resource acquisition, and control
65%VerifiedThe capability gap between frontier AI models is narrowing, with major AI labs rapidly catching up with each other on key capability dimensions.
65%VerifiedIn recent years, key safety researchers at organizations including OpenAI and Google DeepMind have left for similar reasons.
65%UnverifiedAI对齐问题的研究可以追溯到2014年Nick Bostrom的《超级智能》一书
85%UnverifiedUsing AI to supervise AI has a fundamental paradox: it is essentially the undecidable Turing halting problem, which is the core reason AI safety cannot be fundamentally resolved.
70%UnverifiedThe core controversy surrounding the concept of superintelligence lies in the control problem—namely, whether humans can maintain effective constraints once machine intelligence surpasses human intelligence.
60%UnverifiedMany people in the effective altruism community believe that superintelligent AI could pose an existential-level risk to humanity.
60%UnverifiedWhen a system smarter than you appears, you may not even realize it is working against you; this is not an arms race in the traditional sense.
50%UnverifiedCompetition surrounding superintelligence technology has spread from the academic and commercial spheres to the national security level, and AI companies must not only defend against external hackers but also guard against organized infiltration by insiders.
50%UnverifiedBostrom's 2014 book Superintelligence raised the 'alignment problem': how to ensure that a system far smarter than humans always has its goals aligned with what humans truly want.
50%