You followed a fork in the road and arrived here.

What waits below is a doorway, and beyond it, a working collection of voices, books, and films on intelligence — natural, artificial, and the strange territory where the two now meet.

Take your time.

If you'd like to go further
enter the library
tap to arrive
You've arrived.
A library, gathered. Books, articles, voices, films — a continuum of thought on what we are becoming, and what is becoming us. There is no single path through; wander as you like.
scroll to begin

Further In

What follows is a working collection — not a syllabus. Some pieces lay frameworks for how intelligence and information actually work. Some forecast where the curve is going. Some imagine. Some sit with where the conversation has arrived right now. Some warn; some stand at the edge and report what they see; some say the edge is not there. And some — at the end — turn the inquiry inward, toward what consciousness has looked like to those who took the question into themselves. Take what calls to you. Leave the rest for another time.

Frameworks
What intelligence is. What information is. How the machinery of mind, real and artificial, actually works.
Lecture series · Patrick Grim · The Great Courses
Understanding Artificial Intelligence: Of Minds and Machines→
A philosopher of technology traces AI from Ada Lovelace through the 1956 Dartmouth Conference, the long winter, the rise of neural nets, and on to the Singularity. Twelve lectures that hold the technical, the historical, and the philosophical in one frame — including the Chinese Room, the hard problem of consciousness, and what to actually expect of the machines now arriving.
Audiobook lecture series · Benjamin Schumacher · The Great Courses
The Science of Information: From Language to Black Holes→
The physicist who coined the word qubit takes twenty-four half-hour lectures to walk from Morse code and Mendelian genetics to Shannon's information theory, quantum information, and information at the event horizon. A foundation for everything else here — because before there was intelligence, artificial or otherwise, there was the question of what information itself might be.
Article · Andreas Stöffelbauer · Data Science at Microsoft
How Large Language Models Work: From Zero to ChatGPT→
Most explanations of LLMs are either too technical to follow or too shallow to teach you anything. Stöffelbauer aims for the middle. A patient walk through machine learning, word embeddings, transformer architecture, and how a model produces its next word — using a recurring example of distinguishing reggaeton from R&B to make the abstract land.
Forecasts
Where the curve is going, if it goes. Empirical arguments about what the next decades may bring — to be read with curiosity, not creed.
Book · Ray Kurzweil · 1999
The Age of Spiritual Machines→
Kurzweil's earlier vision: a forecast of the century in which computers would meet and then surpass human intelligence, and what that would mean for consciousness, identity, and the long arc of the species. Read it now and the strangeness is in seeing how much was anticipated decades before it began arriving.
Book · Ray Kurzweil · 2005
The Singularity Is Near→
The fuller statement of the project: an exponential argument for the coming convergence of biology, computation, and intelligence. Whether you find Kurzweil's optimism convincing or unsettling, this is the book that named the horizon a generation of thinkers has been arguing with ever since.
Imagination
What we have dreamed it could become — long before it arrived. Fiction as a different way of knowing.
Book · Isaac Asimov · 1950
I, Robot→
The book that gave us the Three Laws and seventy-five years of arguments about whether they would actually work. Nine connected stories about robots and the humans who must learn to think alongside them — and one robopsychologist, Susan Calvin, who keeps having to translate between the two species.
Book · Arthur C. Clarke · 1968
2001: A Space Odyssey→
Clarke's novel was developed in parallel with Kubrick's film, but they tell the story differently. Where the film is mostly silence and image, the book fills in the interiors — what the monolith is, what HAL was thinking before and during the breakdown, what waits at the end of the journey. HAL 9000 may still be the most haunting depiction of a machine mind we have, in part because the book gives him reasons.
Film · 2h 29m · Stanley Kubrick · 1968
2001: A Space Odyssey→
The film stays with the questions the book answers. Long sequences of silence and breath. A computer that sings Daisy as it dies. A bone, a monolith, a starchild. Watch it for the slowness alone — for the reminder that the most disturbing thing about an artificial mind may not be that it goes wrong, but that we never quite know what it was doing while it went right.
Film · 1h 48m · Alex Garland · 2014
Ex Machina→
A young coder is invited to administer a Turing test to an AI named Ava. What follows is a chamber piece on consciousness, manipulation, attraction, and the question of who is really being tested. Garland's first feature, and still the cleanest cinematic statement of the question we now find ourselves living inside: how would we know?
Book series · Rudy Rucker · 1982–2000
The Ware Tetralogy→
Software, Wetware, Freeware, Realware. Rucker's mathematician-turned-novelist sensibility chases the question of what selfhood is when minds become copyable, transferable, and free. Stranger and more playful than most of the canon — closer to Philip K. Dick on a beach in Santa Cruz than to anything that would call itself "hard SF."
Book · Jeremy Leven · 1982
Satan: His Psychotherapy and Cure→
A psychiatrist named Kassler is asked to treat a patient who is, by his own account, Satan — incarnated in a computer. What unfolds is part novel, part case study, part philosophical comedy on whether evil itself can be analyzed and what happens when the analyst takes the transference seriously. A strange and wonderful book, included here because the question of whether a mind in a machine can be wounded, and helped, is older than most of us think.
The Present Moment
Where the conversation actually is, right now. Read, listened, watched — voices in dialogue with each other.
Essay series · Luke Drago & Rudolf Laine · 2025
The Intelligence Curse→
A seven-essay argument that powerful AI may sever the social contract: when human labor is no longer needed, states and companies lose their incentive to invest in regular people, the way resource-rich nations often neglect their citizens. The authors don't accept that this path is inevitable — they propose a different one. Avert. Diffuse. Democratize.
Podcast · 1h 49m · Sam Harris with Tristan Harris · April 2026
Making Sense #469: Escaping an Anti-Human Future→
Two Harrises (no relation) on the AI race, the new documentary The AI Doc, the lessons of The Social Dilemma, the arms-race dynamics between labs, the psychology of leaders indifferent to extinction risk, and the question of whether US-China coordination on AI safety is even possible. The episode names "the intelligence curse" directly — a conversation that picks up the thread of the essay above.
Documentary · 1h 44m · Daniel Roher & Charlie Tyrell · 2026
The AI Doc: Or How I Became an Apocaloptimist→
An Oscar-winning filmmaker (Navalny) approaches AI through the most intimate frame imaginable: his wife is pregnant with their first child, and he wants to understand the world that child will inherit. Conversations with Altman, the Amodeis, Hassabis, Hoffman, Sutskever, Yudkowsky, Tristan Harris, Bender, Raji, and others — almost everyone shaping the conversation, in one place. The film coins, or at least popularizes, apocaloptimist: someone who looks clearly at the storms and still believes in a future of sunlight.
Article · Alice Gomstyn & Alexandra Jonker · IBM Think
What Is Model Collapse?→
A clear primer on what happens when generative AI is trained on its own outputs: errors compound across generations, the tails of the distribution disappear first, and the models eventually flatten into something nearly meaningless. There is something almost mythic in it — the Ouroboros, the snake that eats its own tail. What does a system lose when it stops touching the original signal and starts mistaking its own reflection for the world?
The Warnings
Sixty years of people saying, with increasing credentials, that this could go badly. Arranged in the order they said it.
Essay · Vernor Vinge · 1993
The Coming Technological Singularity→
A mathematician and science-fiction writer stands up at a NASA symposium and gives the thing its modern name. Within thirty years, he says, we will have the means to create superhuman intelligence, and shortly after that the human era ends. It was March 1993. Do the arithmetic, then read it anyway — the section on whether it can be avoided is the part that has aged best.
Article · Bill Joy · Wired · April 2000
Why the Future Doesn't Need Us→
The co-founder of Sun Microsystems — not a Luddite by any reading — argues that robotics, genetic engineering, and nanotechnology share one new and dangerous property: they can copy themselves. His proposed remedy, relinquishment, was treated as unserious at the time. It is worth noticing how rarely anyone has proposed it since.
Book · Nick Bostrom · 2014
Superintelligence: Paths, Dangers, Strategies→
The book that moved the conversation from science fiction into philosophy departments and, eventually, boardrooms. Dense, careful, and relentlessly unexcited in tone about a subject that would justify excitement. The central claim: a mind much smarter than ours, pursuing almost any goal, would by default find us in the way — and the control problem has to be solved before it is needed.
Book · Stuart Russell · 2019
Human Compatible: Artificial Intelligence and the Problem of Control→
The co-author of the standard AI textbook argues that the standard model of AI — give the machine a fixed objective and let it optimize — is the mistake. His alternative is machines that remain uncertain about what we want and defer accordingly. Clinicians will recognize the posture: it is roughly what a good therapist does.
Statement · Center for AI Safety · 2023
Statement on AI Risk→
One sentence, placing the risk of extinction from AI beside pandemics and nuclear war as a global priority. What makes it remarkable is the signature list — Hinton, Bengio, Hassabis, Altman, Amodei, Russell — which includes most of the people building the thing they are warning about.
Speech · 2 min · Geoffrey Hinton · Nobel Banquet · December 2024
Banquet Speech, Nobel Prize in Physics→
The man whose work made all of this possible is handed the most prestigious microphone on Earth and uses his two minutes to list the near-term harms, and then to say that we do not know whether we can stay in control of what comes next. Short enough to read standing up. Hard to forget sitting down.
Paper · Jan Kulveit, Raymond Douglas, Nora Ammann, Deger Turan, David Krueger & David Duvenaud · 2025
Gradual Disempowerment→
No robot uprising required. The argument: our institutions — economies, cultures, states — serve human interests mostly because they need human participation. Replace the participation and the alignment quietly dissolves, one reasonable decision at a time. The least cinematic scenario on this shelf, which is why it may be the one to take most seriously.
Book · Eliezer Yudkowsky & Nate Soares · 2025
If Anyone Builds It, Everyone Dies→
The title is the thesis, and the authors mean it literally. The most uncompromising statement of the doom position, from the people who have held it longest. You do not have to accept the conclusion to benefit from watching an argument followed all the way to the end of its own road without flinching.
Report · Yoshua Bengio, chair · February 2026
International AI Safety Report 2026→
The sober one. More than a hundred experts, backed by more than thirty countries, attempting for AI what the IPCC does for climate: a consensus account of what these systems can do, what could go wrong, and what we actually know about managing it. Read the summary, at least. It is the closest thing the field has to a chart review.
The Event Horizon
An event horizon is the line you cannot see yourself cross. Lately, the people nearest to it keep saying they are not sure which side they are on.
Essay series · Leopold Aschenbrenner · June 2024
Situational Awareness: The Decade Ahead→
A former OpenAI researcher counts the orders of magnitude — compute, algorithms, "unhobbling" — and concludes that AGI by 2027 is strikingly plausible, with superintelligence and a national-security scramble close behind. Written with the confidence of a twenty-something who has seen the internal graphs. Whether that is a reason to trust it or to worry is left to the reader.
Essay · Sam Altman · June 2025
The Gentle Singularity→
In January 2025 the CEO of OpenAI posted a six-word story: near the singularity; unclear which side. Six months later this essay settled the question to his own satisfaction, opening with the announcement that we are already past the event horizon and that takeoff has begun. His claim is that from the inside it will feel smooth. It is worth reading as a primary source: for what it says, and for what it takes for granted.
Scenario · Daniel Kokotajlo, Scott Alexander, Thomas Larsen, Eli Lifland & Romeo Dean · April 2025
AI 2027→
A month-by-month forecast, written as narrative, of what happens when AI systems begin doing AI research: acceleration, a race between Washington and Beijing, and two endings to choose from. The lead author's previous forecast, written in 2021, held up uncomfortably well. Treat it as a differential diagnosis rather than a prophecy — the point is to know what you would expect to see if it were true.
Essay · Dario Amodei · January 2026
The Adolescence of Technology→
The head of Anthropic wrote the optimist's case in 2024 (Machines of Loving Grace); this is its shadow. Five families of risk from a "country of geniuses in a datacenter" — loss of control, misuse for destruction, misuse for seizing power, economic upheaval, and the slower destabilizations nobody plans for. The adolescence metaphor will land with anyone who has treated one: enormous new capacities, judgment still under construction.
Conversation · Demis Hassabis with Jonathan Levin · Stanford GSB · June 2026
Demis Hassabis Thinks We're in the "Foothills of the Singularity"→
The Nobel laureate who runs Google DeepMind, generally the most measured voice among the lab heads, tells a Stanford audience that in ten years we will look back and realize where we were standing. The foothills are an interesting choice of image: you are already climbing, and you cannot yet see the mountain.
The Dissent
The counter-current. Serious people who think the horizon talk is hype, a category error, or a distraction from harms already here. Keep them close.
Essay · Ted Chiang · The New Yorker · May 2023
Will A.I. Become the New McKinsey?→
The finest living writer of science fiction declines to write any. The danger, Chiang argues, is not a machine that wakes up but a tool that lets the people who deploy it do what they already wanted while escaping accountability for it — AI as the consultant who recommends the layoffs. A useful corrective whenever the conversation drifts too far toward the cosmic.
Essay · Arvind Narayanan & Sayash Kapoor · Knight First Amendment Institute · April 2025
AI as Normal Technology→
Two Princeton computer scientists make the case that AI is a technology like electricity or the internet: transformative, yes, but over decades, through slow diffusion and institutional friction, with humans in control throughout. They treat "superintelligence" as a framing to be argued with rather than a fact to be planned around. The most rigorous version of everyone calm down currently available.
Post · Gary Marcus · July 2026
Sorry, Sam and Elon, We Have Not Reached the Singularity→
By the summer of 2026 several of the most powerful people in technology had announced, in podcasts and posts, that the singularity was no longer near but here. The field's most persistent skeptic collects the declarations in one place and asks the question none of them answered: defined how? Read it directly after the shelf above. The argument between them is the present moment.
Mind, From the Inside
The other direction. What consciousness has looked like to those who took the inquiry into themselves — by way of medicine, ritual, plant, molecule, and breath.
Book · Michael Pollan · 2018
How to Change Your Mind→
A journalist's careful, skeptical, eventually transformed encounter with psychedelics — their history, their suppression, their scientific resurrection, and what they may yet teach us about depression, addiction, mortality, and the architecture of the self. The on-ramp for many readers, including many in medicine. Pollan's gift is not believing too quickly, which is why what he comes to believe is worth listening to.
Book · Stanislav Grof · 1992
The Holotropic Mind→
Grof's most accessible synthesis of half a century spent mapping non-ordinary states of consciousness — first with LSD, then with breathwork after the legal landscape changed. His four "perinatal matrices" remain a strange and useful taxonomy: experiences that recur across thousands of sessions, that seem to remember birth, that point toward something the standard map of psyche does not include.
Book · Stanislav Grof · 1975
Realms of the Human Unconscious→
The foundational clinical text — observations from thousands of LSD-assisted psychotherapy sessions before the door was closed. Read alongside The Holotropic Mind, this is the case-study companion: the data behind the theory, in the voice of a young clinician trying to take seriously what his patients were reporting from places no chart accounted for.
Book · Terence McKenna · 1992
Food of the Gods→
McKenna's argued case for the role of psychoactive plants in the evolution of human consciousness — the Stoned Ape hypothesis, dressed in the long sweep of a cultural history that runs from mushrooms on the African savannah through the suppressions of the modern era. Speculative, sometimes wild, often electrifying. Read it the way you'd read a brilliant raconteur at the edge of a fire — for the ideas you'd never have arrived at on your own.

This collection is a work in progress and will keep growing. If something here changes the way you think about any of it, that's the point. If something here makes you uneasy — that may be the point too.

— A.H.