ai

Robert Miles explains instrumental convergence: why an AGI with almost any goal would tend to pursue sub-goals like self-preservation and resource acquisition. It builds on Omohundro's basic AI drives and his orthogonality thesis video.

Robert Miles examines whether the case for AI safety is a Pascal's mugging, a tiny probability of an enormous payoff. He argues it is not, because AI risk does not depend on vanishingly small probabilities.

Robert Miles introduces the first of several videos on the paper 'Concrete Problems in AI Safety', focusing on how AI systems can cause accidental negative side effects. It explains the problem in accessible terms.

Robert Miles goes through ten common reasons people give for dismissing AI safety, adapted from Stuart Russell's list, and answers each one. It is aimed at skeptics and newcomers.

An MTA video on communications-based train control, the system meant to replace the New York City subway's century-old signaling. It also shows the mechanical interlocks and signals in use today.

CGP Grey explains split-brain patients, where severing the corpus callosum lets the two hemispheres act independently. He uses this to explore the idea of two minds in one head.

Wendover Productions explains how airlines make money from different cabin classes and why premium seats matter so much to profits. It also touches on how this shaped decisions such as the decline of supersonic travel.

3Blue1Brown shows how the uncertainty principle arises from Fourier transforms. A wave packet cannot be narrow in both position and frequency, so the principle is a property of waves rather than of measurement.

Wendover Productions explains why US passenger rail lags behind other countries. It covers freight companies owning most of the track, geography and population density, and the economics of competing with flying.

A Wendover Productions explainer on why train tickets, especially Amtrak in the US, cost so much compared with flying. It breaks down staffing, speed and the cost components behind a ticket.

A Wendover Productions explainer on how China built the world's largest high-speed rail network so quickly and why its rail strategy differs from America's. It covers politics, social benefits and the economics of unprofitable lines.

A Wendover Productions explainer arguing that major airlines make much of their profit from loyalty programs and co-branded credit cards rather than flying. It frames airlines as quasi-banks that sell points in bulk, using United's pandemic-era loan as a hook.

Dave Plummer tours the IBM z16 mainframe from design through factory assembly, test lab, accelerators and water cooling. He explains why mainframes still exist and what makes them reliable.

Wendover Productions explains how UK rail privatization worked and why it failed. It covers the fragmented operator and infrastructure structure and why market forces apply poorly to rail.

A Wendover Productions explainer on how airport security is designed, covering screening layers, x-ray scanning, and the contrast with Israel's profiling-based approach. It walks through the logic behind why security works the way it does.

Kurzgesagt examines how AI-generated content is flooding the internet, from video and music to news and books. It looks at how this makes verifying what is real harder.

A Wendover Productions explainer on how budget airlines can sell tickets for under 10 euros. It covers the costs airlines face and how low-cost carriers structure their business.

Musician Benn Jordan uses AI to detect AI-generated music and digs into music industry data, including streaming payouts and sample library copyright problems. He covers ways to tell whether a track is human-made.

A short clip of Richard Feynman explaining how a train stays on its tracks, through the conical shape of wheels and how it handles curves without differentials.

Not Just Bikes examines Toronto's GO Transit commuter rail, arguing it is a costly system with park-and-ride stations and rush-hour-only service that subsidizes car-dependent suburbs.

A short clip from the 1983 BBC series Fun to Imagine in which Richard Feynman explains how a train stays on the track. He covers the geometry of tapered wheels that steer the train back toward the center of the rails.

Practical Engineering explains how continuous welded rail copes with thermal expansion by restraining it, using the relationship between temperature, stress and strain. The video is part of a series on railways.

CGP Grey explains the history and legal status of territorial claims on Antarctica. It covers the overlapping national claims and the Antarctic Treaty that freezes them.

Hugh Jeffreys tears down the iPhone 12 and assesses its repairability, focusing on how parts are serialized and tied to the device so swapping components between phones does not work.

An Economics Explained video on the different kinds of billionaires, from founders to oligarchs. It explains how wealth is held in illiquid stock and why diversification is a problem for some.

RealLifeLore explains with maps why Russia invaded Ukraine, focusing on energy resources, security buffers, Black Sea access and Crimea's strategic value.

Technology Connections explains why single-hose portable air conditioners are inefficient and often a poor choice compared with window units. It details how the single hose pulls in warm outside air and wastes energy.

Louis Rossmann gives a basic, non-technical introduction to right to repair legislation for people new to the issue.

An RSA Animate summary of Iain McGilchrist's lecture on the divided brain, arguing the brain's two hemispheres attend to the world differently and that this has shaped culture and society.

Brian Douglas introduces Braitenberg vehicles, simple robots whose sensors are wired directly to motors yet show surprisingly lifelike behavior. He draws on Valentino Braitenberg's book Vehicles.

Half as Interesting explains why airlines flew empty 'ghost flights' out of Heathrow to keep their landing slots under use-it-or-lose-it rules.

Destin Sandlin of Smarter Every Day explores how grain bins work and the engineering and know-how involved in farming. His point is that farmers are exceptionally capable.

Robert Miles explains the orthogonality thesis: how an agent can be highly intelligent while pursuing goals that seem stupid or alien to humans. He separates intelligence from goals and touches on the stamp-collector example.

Robert Miles asks whether corporations can be thought of as artificial general superintelligences and how far the analogy holds. He argues the differences, such as being limited by employees' capabilities, matter for AI safety.

Robert Miles explains reward modelling, in which an AI learns what humans want from their preference comparisons instead of a hand-written reward function. He walks through the 2017 paper on deep reinforcement learning from human preferences.

Robert Miles covers a paper where researchers ran real versions of his earlier mesa-optimizer thought experiments. They show inner misalignment, where an AI learns the wrong goal in training and misbehaves when deployed.

A Computerphile video in which Rob Miles warns against assuming a general AI will think like a human. He uses the thought experiment of a stamp-collecting AI to show how a goal-driven system could pursue its aim with dangerous results.

A Veritasium video on AlphaFold, DeepMind's protein structure prediction system, and the protein design work it enabled. It features interviews with John Jumper and with David Baker's protein design group and explains why predicting protein folding matters.

Tom Scott features Kurtis Baute inside a sealed homemade biodome as carbon dioxide builds up, showing how stale air with elevated CO2 affects cognitive function. Sources include studies on CO2 and cognitive scores.

Rob Miles explains on Computerphile why implementing an off switch for a general AI is hard. A capable agent would resist being stopped because shutdown prevents it from achieving its goal.