mental models

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.

Rob Miles explains on Computerphile how language models generate plausible text and how transformers improved on recurrent networks, with a look at what comes next.

A Technology Connections video essay on how analog and digital clocks shape different mental models of time, with the minute hand acting as a progress indicator.

A short cartoon explaining how language models like ChatGPT work, from autocomplete and conditional probabilities to neural networks, gradient descent and backpropagation. It shows how networks act as universal function approximators.

Roaring Kitty's July 2020 analysis of GameStop, arguing a bullish deep-value case when the stock was about $4 and short interest exceeded 100%. It covers fundamentals, cash balance and technicals.

A Kurzgesagt animated explainer on the climate and environmental cost of meat. It covers how beef drives deforestation and emissions, and how this compares with plant-based foods.

Sabine Hossenfelder explains climate sensitivity and why recent observations suggest climate models may have underestimated warming. She says this changed her view on how worried to be.

Rob Miles discusses GPT-3 on Computerphile, covering how scaling parameters kept improving results over GPT-2 and how the model nearly learned arithmetic without anyone knowing exactly how.

3Blue1Brown's visual explanation of how transformers and GPT-style language models work, tracing data through tokens, embeddings and attention.

Veritasium's explainer on Godel's incompleteness theorems and the limits of mathematical proof, tracing set theory, Hilbert's program and the road to Turing and modern computing.

Matt Parker of Stand-up Maths explains why people keep making the same mistake with millions and billions, using a Bloomberg cash calculator example. He looks at the unit slip behind the repeated arithmetic error.

A four-minute minutephysics video arguing that causal models and causal networks let statistics infer causation from correlation. It touches on Reichenbach's principle, common causes, feedback and quantum causal discovery.

A guest video by Welch Labs on 3Blue1Brown explaining how text-to-image and video models work. It covers CLIP, shared embedding spaces, diffusion models (DDPM, DDIM), conditioning, guidance and negative prompts.

Forgotten Weapons examines the Heckler & Koch G11, a West German experimental rifle built around a caseless cartridge. The video covers the G11K2 and its unusual technology before its cancellation for political and economic reasons.

CGP Grey's sarcastic animated video adapted from the book How to Be Miserable, listing habits that reliably make people unhappy. By inverting the advice, it shows what to avoid and prompts reflection on goals and routines.

Everyday Astronaut's Tim Dodd runs an hour-long, data-driven analysis of what rockets actually emit and how that compares with airliners and other industries. It breaks down exhaust by propellant type, per-passenger and total emissions, and ways launches could be cleaner.

Veritasium tells the story of Thomas Midgley Jr., who introduced leaded gasoline and CFCs, and covers the science behind lead exposure and ozone damage. It also touches on isotope geochemistry used to uncover the history.

Kurzgesagt explains why humans feel loneliness, its evolutionary role, and how modern life can turn it into a harmful downward spiral. It cites research and offers ideas on what to do about it.

Huygens Optics explores the idea that space is a non-linear elastic medium, and uses simulations to show how wave energy might be confined locally into particle-like objects. It touches on inertia, the Schwinger limit and the vacuum as a medium.

Dr. K of HealthyGamerGG argues that gifted kids can be seen as special needs, tracing a path from the gifted label to burnout and perfectionism.

Alex O'Connor shows that ChatGPT cannot draw a wine glass filled to the brim, and ties the failure to David Hume's missing shade of blue. It asks what the failure says about how image models learn.

Dr. K of HealthyGamerGG explains why people feel unmotivated and unable to start tasks after hours of YouTube or similar distractions. He frames it as a result of how social media is designed to capture attention, and says not to blame yourself.

Computerphile video in which Mike Pound explains a University of Maryland paper on watermarking LLM output. The method biases token choices so generated text can later be detected statistically.

Veritasium visits Richard Lenski's lab to explain the decades-long E. coli evolution experiment, which asks how repeatable evolution is and shows bacteria evolving to use a new food source.

CGP Grey's video "Humans Need Not Apply" argues that automation will make a large part of the workforce economically obsolete, much as engines did for horses. It surveys robots and software replacing both physical and mental labor.

Joseph Everett, with Dr. Frank Mitloehner, argues that eating less meat will not do much for the climate. He covers beef's water footprint, land and feed use, and methane.

Not Just Bikes explains how many American suburbs, built all at once rather than incrementally, cannot cover their infrastructure costs and survive on growing debt. It draws on Strong Towns' work on the growth Ponzi scheme.

3Blue1Brown builds toy SIR models to simulate how an epidemic spreads and tests scenarios like social distancing, travel restrictions, central gathering places and quarantine. It was made during the early COVID-19 outbreak.

A 1936 Chevrolet-sponsored film that explains how a car transmission works by treating gears as spinning levers. It uses physical models to make the mechanism clear.

Guitarist Jim Lill methodically tests what in an electric guitar affects its tone, swapping parts on a partscaster and even trying a 2x4 body. He finds that wood and construction matter far less than people assume.

A snarky, comedic video essay from Climate Town, with a guest appearance by Not Just Bikes, arguing that American suburban sprawl and single-family zoning are financially and environmentally unsustainable. It covers how zoning rules shape housing and what viewers can do locally, such as attending zoning meetings.

A PBS Space Time episode hosted by Gabe Perez-Giz on the nature of spacetime and whether space and time are fundamental. It builds up reference frames and causality in a short, dense format.

Arthur Keller gives a roughly half-hour crash course on environmental collapse, covering exponential growth, carrying capacity and ecological overshoot.

A Welch Labs explainer on neural scaling laws and the compute-efficient frontier, asking why model loss seems to hit a line it cannot cross. It draws an analogy to an ideal gas law for AI.

A Kurzgesagt explainer asking whether organic food is really healthier or better for the environment. It defines what organic means and looks at pesticide use, nutrition and environmental impact, with sources.

A School of Life short introducing Michel Foucault, who questioned assumed progress in how society treats madness, medicine and sexuality.

A Kurzgesagt animated explainer on meat: why people love it and what it costs animals, the environment and health. It lays out the complicated trade-offs of meat production and consumption.

A five-minute animated explainer of liquid democracy, in which voters can delegate their vote to trusted people. It starts from the origins of democracy and contrasts direct and representative models.

A short explainer from Nat and Friends on where Google Earth's 3D imagery comes from, how it is stitched together, and how often it is updated. It describes aircraft carrying camera rigs and photogrammetry that turns the photos into 3D models.

PBS Space Time examines Roger Penrose's idea that consciousness arises from quantum processes in the brain, and reviews new evidence about microtubules. It ends with an estimate of what replicating consciousness would demand.