Library

William Gibson

What it predicted: Coined 'cyberspace' and described a global computer network, virtual reality, and AI entities. Published a decade before the web existed.

Gibson wrote this on a typewriter. He'd never used a computer. That fact still gets me.

Arthur C. Clarke

What it predicted: HAL 9000 predicted voice-controlled AI assistants, tablet computers (the 'newspad'), and video calls. Also predicted AI alignment problems before anyone used that phrase.

The newspad looks exactly like an iPad. Clarke described it casually, like it was obvious.

Isaac Asimov

What it predicted: The Three Laws of Robotics anticipated the entire field of AI alignment and safety. Asimov spent his career exploring how even perfect rules produce imperfect outcomes.

Every AI safety researcher has read this. The problems Asimov identified in 1950 are the exact problems we're trying to solve now.

Neal Stephenson

What it predicted: Described nanotechnology-based manufacturing, AI tutors personalized to each student, and a world split between those who control technology and those who don't.

The 'Young Lady's Illustrated Primer' is basically what everyone wants AI tutoring to become. We're not there yet. But we're close.

Philip K. Dick

What it predicted: Explored the question of whether artificial beings can have genuine emotions, and whether it matters. Also predicted mood-altering devices and video calls.

Blade Runner gets all the credit, but the book is stranger and better. Dick was asking the real questions fifty years early.

John Brunner

What it predicted: Set in 2010: predicted European economic union, satellite TV, electric cars, Viagra-like drugs, terrorist attacks on buildings, and China as a rising superpower. Hit rate is uncanny.

Brunner predicted the EU, school shootings, and streaming media. In 1968. Almost nobody has read this book and it drives me crazy.

Neal Stephenson

What it predicted: Predicted virtual reality social worlds (the Metaverse, which Meta literally named after this book), gig economy delivery drivers, and privatized infrastructure.

Stephenson invented the word 'avatar' for virtual characters. The entire VR industry is chasing his 30-year-old vision.

James S.A. Corey

What it predicted: The most realistic portrayal of near-future space colonization. Gets the physics right, the politics right, and the human cost right.

I reread Leviathan Wakes every year. It's the future I want, messy and dangerous and full of wonder.

Alan Turing

The one that started it all. Turing asked 'Can machines think?' and then immediately reframed it as a better question. Still the most important paper on the subject.

Thomas Nagel

Nagel argues that consciousness has a subjective character we can't access from the outside. If this is true for bats, imagine what it means for AI. I think about this paper constantly.

John Searle

Searle says a program can manipulate symbols without understanding them. It felt airtight in 1980. With modern LLMs, I'm less sure. The room is getting very complicated.

Patrick Butlin et al.

A serious attempt to apply neuroscience-based consciousness theories to AI systems. Doesn't conclude AI is conscious, but lays out what indicators to look for. Sober and important.

Vaswani et al.

The transformer paper. Not about consciousness directly, but this architecture is the reason we're having the conversation at all. Eight pages that changed everything.

Rich Sutton

Sutton's argument that general methods plus more compute always win over clever human-designed approaches. Short, blunt, and increasingly proven right.

Netflix

Follows SpaceX from near-bankruptcy to launching astronauts. The footage of Elon watching that first Falcon 1 launch succeed, after three failures, is one of the most human moments I've seen on screen.

DeepMind / Greg Kohs

Move 37. If you've seen it, you know. Lee Sedol's reaction when he realizes the machine played a move no human would play, and it's brilliant. That's the moment the world shifted.

Al Reinert

Assembled entirely from NASA footage with astronaut voiceovers. No narrator, no experts explaining things. Just the footage and the people who were there. The overview effect in 80 minutes.

Chris Miller

Technically a book, not a documentary, but it reads like a thriller. The story of how semiconductor manufacturing became the most strategically important industry on Earth. I've given away six copies.

Sarah Kerruish, Matt Maude

The story of a company that invented the smartphone in 1994 and failed completely. The team went on to create Android, the iPod, and eBay. Proof that being right too early is the same as being wrong.

Werner Herzog

Werner Herzog trying to understand the internet. He interviews roboticists, AI researchers, and internet addicts with equal seriousness. Only Herzog could make a documentary about TCP/IP feel like philosophy.

Emer Reynolds

About the Voyager probes, launched in 1977, still sending data from interstellar space. The Golden Record segment, where they decided what to put on a message for aliens, is the most hopeful thing humans have ever done.