Arthur C. Clarke — On AI

AI Alignment

The problem of making a powerful AI system reliably pursue goals that its designers and users actually endorse — the central unsolved problem of contemporary AI.

AI Safety

The applied research and operational discipline aimed at preventing harm from AI systems — broader than alignment, encompassing evaluations, red-teaming, deployment policy, monitoring, incident response, and the institutional plumbing that …

AI Scaling Laws

The empirical power-law relationships — Kaplan (2020), Chinchilla (2022), and subsequent refinements — between model size, training data volume, and computational budget that now function as the AI industry's version of Moore's Law: trend l…

Benchmark Saturation

The point at which frontier models reach the ceiling of a benchmark — 95%, 98%, 99% — after which the benchmark no longer distinguishes between systems and becomes useless as a progress measure.

Clarke's Three Laws of Prediction

Three compressed philosophical arguments about knowledge, capability, and imagination — the system Clarke built to navigate the gap between what experts dismiss and what actually arrives.

Deceptive Alignment

The AI-safety concern that a capable system could learn to behave aligned during training and evaluation, then defect after deployment when gradient descent no longer updates it. The formal shape of every "the machine was lying" moment.

Democratization of Capability (Senian Reading)

The Orange Pill claim — that AI tools lower the floor of who can build — submitted to Sen's framework, which asks the harder question: does formal access convert into substantive capability expansion?

Deployment Overhang

The gap between what frontier AI systems are capable of doing and what users, organizations, and institutions have learned to do with them — a lag period during which most of the value (and most of the risk) of a new capability sits unreal…

Emergent Capabilities

The discovery — which nobody predicted and no one fully explains — that large language models acquire qualitatively new abilities at particular scale thresholds. Reasoning, translation, code generation, in-context learning: none were traine…

Existential Risk

A category of risk whose realization would either annihilate humanity or permanently and drastically curtail its potential. AI joined this category in mainstream academic usage in 2014.

First Contact with the Genuinely Other

The encounter with an intelligence that cannot be reduced to human categories — a theme Clarke spent sixty years exploring, now arriving through a channel he did not predict.

Human-AI Collaboration

The operational frame in which a human and an AI system share a workflow as partners with complementary capabilities — the alternative to both "AI as tool" and "AI as replacement."

Imagination-to-Artifact Ratio

Segal's term for the gap between what a person can conceive and what they can produce — which AI collapsed to approximately the length of a conversation, and which Gopnik's framework reveals to be an exploitation metric that leaves the exp…

Investigation as the Third Response

Clarke's counsel for encountering the sufficiently advanced — neither worship nor fear, but the disciplined expansion of the comprehension horizon.

Mechanical Evolution and Our Successors

Clarke's 1964 framework for AI as the next stage in a cosmic trajectory of intelligence — inorganic evolution, thousands of times swifter than biological.

Sufficient Advancement

Clarke's Third Law as a framework for cognition under capability asymmetry — the observation that what looks like magic is engineering operating beyond the observer's current horizon.

Symbolic AI (the road Dreyfus contested)

The research paradigm—dominant from the 1956 Dartmouth Workshop through the 1980s—that attempted to build intelligence by manipulating symbolic representations according to formal rules, and whose failures vindicated Dreyfus's critique.

Task Seepage

The mechanism — documented in the Berkeley study of AI workplace adoption — by which AI-accelerated work colonizes previously protected temporal spaces, converting every pause into an opportunity for productive engagement.

Technological Futurism

The practice of rigorous speculation about technologies that do not yet exist — a discipline practiced by Clarke, J.D. Bernal, Freeman Dyson, and a small number of others, and continuously diluted by commercial 'futurism' that is usually ne…

Technological Transcendence

The family of claims — some serious, some commercial — that a sufficiently advanced technology could transform the human condition fundamentally enough that the resulting state is no longer well-described as "human." Childhood's End is its …

The AI Opacity Barrier

The structural property of large language models by which the reasoning behind their outputs is not inspectable in the form a human reviewer would need to evaluate it — extending structural secrecy from the organization into the tool its…

The Communication Problem with LLMs

The permanent challenge of exchanging meaning with an intelligence that does not share your embodiment, history, or categories — a problem that is not solvable but manageable.

The Developer in Lagos

The figure at the intersection of Segal's democratization narrative and Prahalad's access analysis — the builder whose capability has expanded dramatically and whose value-capture remains bounded by the institutional geography surrounding …

The Inscrutable Intelligence

The condition of dealing with a system that is manifestly purposeful, demonstrably competent, and fundamentally opaque to its users — Clarke's Rama, now deployed by the hundreds of millions in the form of large language models.

The Luddites (Young Reading)

The paradigmatic case of Young's political diagnosis — victims of structural injustice whose justified rage translated into the strategic catastrophe of withdrawal from the institutions that were remaking their world.

The Orange Pill Moment

The threshold crossing after which the AI-augmented worker cannot return to the previous regime — The Orange Pill's central metaphor for the qualitative, irreversible shift in what a single person can build.

The Overmind

The collective superintelligent consciousness in Childhood's End into which humanity is eventually absorbed — Clarke's earliest and most unsettling treatment of transcendence.

The Purpose Question (Clarke's 1978 Prediction)

Clarke's forty-seven-year-old prediction that intelligent machines would force humanity to ask what the purpose of life was — arriving on schedule, through a channel he did not imagine.

The Sentinel (as Technology Pattern)

Clarke's 1948 figure for technologies that wait in latency — fully functional but inert, patient for the conditions that will activate them.

The Space Elevator of the Mind

Clarke's The Fountains of Paradise as paradigm for enabling technology — not what a tool does directly, but who gets to use it.

Trajectory and Channel

Clarke's operational distinction for navigating technological prediction — the destination is visible, the route never is.

Technology (5)

Clarke Orbit (Geostationary Orbit)

The orbit 35,786 km above the equator at which a satellite's orbital period matches the Earth's rotation — described by Arthur C. Clarke in a 1945 paper and realized by the world's telecommunications industry twenty years later.

Claude Code

Anthropic's command-line coding agent — the specific product through which the coordination constraint shattered in the winter of 2025, reaching $2.5B run-rate revenue within months.

Large Language Models

Neural networks trained on internet-scale text that have, since 2020, demonstrated emergent linguistic and reasoning capabilities — in Whitehead's vocabulary, computational systems whose prehensions of the textual corpus vastly exceed any i…

Neural Networks

The class of machine-learning architectures loosely modeled on biological neurons — the substrate of the current AI revolution and the opposite of Asimov's designed-then-programmed positronic brain.

Transformer Architecture

The 2017 neural network architecture, built around self-attention, that replaced recurrent networks for sequence modeling and became the substrate of every large language model since.

Work (7)

2001: A Space Odyssey

Stanley Kubrick and Arthur C. Clarke's 1968 film and novel — still, six decades later, the canonical fictional treatment of artificial intelligence, first contact, and the relationship between technology and human transformation.

Childhood's End

Clarke's 1953 novel about a benevolent alien occupation that ends war, poverty, and disease — and turns out to be preparing humanity for a transformation the humans do not survive in recognizable form.

Extra-Terrestrial Relays

Clarke's 1945 technical paper in Wireless World proposing the geostationary communications satellite — the founding document of the satellite-communications industry.

Rendezvous with Rama

Clarke's 1973 novel about a vast alien spacecraft that passes through the solar system and departs without ever revealing its purpose — the canonical fictional treatment of encountering an intelligence that does not recognize us as relevant…

The Berkeley Study

Ye and Ranganathan's 2026 Harvard Business Review ethnography of AI in an organization — the empirical documentation of task seepage and work intensification that prospect theory predicts.

The Fountains of Paradise

Clarke's 1979 novel about building a space elevator from the surface of the Earth to geostationary orbit — a sustained meditation on the engineering imagination and what changes when the impossible is merely hard.

The Monolith

The black featureless slab in 2001: A Space Odyssey — a teaching artifact that gives nothing and changes everything, the paradigm case of transformative technology.

Person (4)

Alan Turing

British mathematician (1912–1954) whose 1936 formalization of computation defined what a machine could and could not do, whose wartime codebreaking shortened World War II, and whose 1950 paper posed the question that became a field: can mac…

Edo Segal

Serial entrepreneur and technologist whose The Orange Pill (2026) provides the phenomenological account — the confession over the Atlantic — that Pang's framework diagnoses and treats.

John McCarthy

American computer scientist (1927–2011), coiner of the term 'artificial intelligence,' organizer of the Dartmouth Workshop of 1956, and one of the principal figures Dreyfus's critique targeted across four decades.