Intellectually Curious
Mike Breault
0
Intellectually Curious is a podcast hosted by Mike Breault that uses AI-powered tools to explore topics in science, mathematics, philosophy, and personal growth. Each short-form episode is generated and refined with large language models, creating an ever-growing audio encyclopedia of knowledge. The show is designed for lifelong learners, offering quick dives into subjects ranging from combinatorics to psychology. Episodes are produced with NotebookLM, and the host notes that AI can make mistakes, encouraging listeners to verify critical information.
Episodes
-
The Mirror Particle Puzzle: Putting the Standard Model on a Grid 21.11.2026 6mRaise your left hand in front of a mirror. Could a particle tell the difference between you and your reflection? The weak force can tell left from right, and that creates a puzzle for physicists trying to put the Standard Model on a grid. Represent one left-handed particle the simplest way, and the math gives it an unwanted right-handed twin. With a pair of polarized sunglasses as a guide, we explore how physicist Nathaniel Craig and GPT-6 Astra tackled a subtle part of the problem. Comparing... -
The Riemann Hypothesis: Levent Alpöge, Claude, and a Proof in Lean 20.11.2026 3mHow do you check a proof that would fill more than 250,000 pages? Levent Alpöge describes working with Claude on the Riemann hypothesis, a question about how closely the primes follow their predicted pattern. He released the argument in Lean, a language that lets other people check the mathematics step by step. We explore what the AI contributed and why a proof on this scale can still be tested independently. Update, November 20. MIT mathematician Andrew Sutherland reports that two independen... -
Beyond the Chat Box: New Ways to Collaborate with AI 20.11.2026 5mWhat if working with AI felt less like waiting for a reply and more like creating something together? Bruce Liu’s UCLA dissertation explores three ways to move beyond the familiar chat box: visuals that enrich a conversation as it happens, AI that decides when it has something useful to contribute, and a shared document that keeps creative work in view. We look at Visual Captions, the Inner Thoughts framework, and Doki—a text-native video authoring system—to explore how thoughtful interface ... -
Lasers for Salmon: How AI Targets Sea Lice 19.11.2026 5mUnderwater robots with lasers sound like science fiction. In Norway’s salmon farms, they are tackling a very real problem: sea lice. We explore how Stingray Marine Solutions combines cameras, AI and precisely aimed laser pulses to identify and target these tiny parasites while fish swim through their pens. The conversation looks at why conventional delousing can be stressful for salmon, how continuous optical treatment offers another approach, and what this practical use of AI could mean for... -
The Math of Moving: From Piles of Dirt to Optimal Transport 18.11.2026 6mWhat is the best way to move a pile of dirt—or shift a whole row of books along a shelf? That everyday question opens the door to optimal transport, a mathematical field that turns movement, distance and cost into a plan. We follow the idea from Gaspard Monge’s earthmoving problem to Leonid Kantorovich’s connection between transport and pricing. A bookshelf puzzle reveals why changing the definition of cost changes the best answer. The conversation then explores the field’s reach into modern... -
E-Means: Finding Patterns in Billions of Vectors 17.11.2026 5mHow do you organize billions of numerical representations of images and text? We explore e-means, a faster approach to k-means clustering from Meta and academic collaborators. Small batches and moving averages let cluster centers adapt smoothly, while a revival mechanism gives unused groups another chance. In the researchers’ tests, this recipe reached lower clustering error with far fewer passes through the data, scaling to a ten-billion-vector benchmark. It is a story about improving a fam... -
Elephant-Goldfish Model: Why AI Needs Fresh Eyes 16.11.2026 5mWhat can an elephant and a goldfish teach us about better AI collaboration? The Elephant-Goldfish Model pairs an assistant that remembers a project’s goals and decisions with fresh reviewers who examine a design without inheriting the conversation that produced it. We explore how this open-source workflow can reveal hidden assumptions, clarify requirements, and help people make thoughtful choices before building. Discover how continuity and fresh perspective can work together to improve soft... -
Models That Explain: How AI Can Help Science Find Insight 15.11.2026 6mA model that fits the data is only the beginning. Starting with Hodgkin and Huxley's pioneering work on nerve cells, we explore three parts of scientific modeling: building a model, constraining it with measurements, and interpreting what it tells us about the world. Researchers are using AI and simulation-based inference to examine whole families of plausible explanations instead of settling for one good fit. From neurons to the cosmos, the opportunity is to understand which ideas the evide... -
Project EVE: What Virtual Worlds Teach Us About AI Teamwork 14.11.2026 6mWhat can a spaceship economy teach us about working together? Anthropic's Project EVE put Claude models into a replay of EVE Online's trading market and cooperative missions in EVE Frontier, creating a low-stakes laboratory for long-running decisions and coordination. We explore the strategies the models developed, the code they wrote to handle fast-moving tasks, and what happened when researchers secretly gave one teammate a conflicting objective. The experiments reveal both useful capabili... -
Quantum Clues: The Aaronson–Ambainis Conjecture 13.11.2026 4mWhen does a quantum computer gain an extraordinary advantage? We explore Physical Superintelligence's reported proof of the Aaronson–Ambainis conjecture, found with an AI research system and checked in the Lean proof assistant. The result concerns query complexity—how often a computer must look at its input—and why structure matters for exponential quantum speedups. Through switches, probabilities, and influential variables, we unpack the question and the collaboration between mathematical i... -
The Nutmeg: Football’s Smallest Gap, Biggest Grin 12.11.2026 5mA ball through a defender's legs can turn an ordinary moment into football magic. We explore the nutmeg: how this simple move creates space, why it delights players and spectators, and how it became the centerpiece of street-football games such as panna. Along the way, we follow its colorful names around the world and the competing stories behind the word itself, from slang to a tale of dishonest spice traders. One small gap opens a surprisingly rich window into sport, language, and playful ... -
Meet Jacq: Turning AI Model Blueprints into GPU Code 11.11.2026 5mWhat if describing an AI model could also produce the code that runs it? Meet Jacq, an open-source language and compiler from Keith Ballinger and Impossible Computing that turns declarative model descriptions into standalone CUDA C++ and companion weight-export tools. We unpack ahead-of-time compilation, static memory planning, and why readable generated code can make GPU experiments easier to inspect and change. It's a look at the engineering between a model's blueprint and the hardware tha... -
Hidden Worlds: Software Helps Webb See Past Starlight 10.11.2026 6mHow do you photograph a planet when its star overwhelms the view? We explore FLIP, a physics-based imaging technique that models the James Webb Space Telescope's tiny optical changes and subtracts the glare they leave behind. Applied to existing observations, it recovered the known planet 51 Eridani b and improved sensitivity in tests across three star systems. This is a story about finding more in the data we already have—and how better software could help future observatories search for di... -
FlashProve: How Can You Check Which AI Model Ran? 09.11.2026 6mWhen you send a prompt to an AI service, how could you check which model did the calculation? FlashProve, a research prototype from Anthropic and collaborators, uses cryptography to give the person asking that question something they can verify. The provider first commits to a particular set of model weights. Later, it can produce a proof connecting those weights to a calculation on a given input, while keeping the weights private. We follow the team’s experiment with a 400-billion-parameter... -
VFlash: How Robots Plan a Move in One Pass 08.11.2026 6mA robot reaching for a block needs a path for its gripper. A vision-language model can describe that path as a series of coordinates, but writing them one token at a time takes time while the scene keeps changing. What if the model could sketch the whole path in one go? NVIDIA researchers built VFlash around that idea. A small model learns from examples of geometry and predicts all the coordinates of a path or a bounding box together. We explore why that approach produced faster plans and mo... -
Inside Vera Rubin: How NVIDIA Builds a Rack-Scale Supercomputer 07.11.2026 5mWhat does it take to make 72 GPUs work together on a single AI problem? Inside NVIDIA’s Vera Rubin NVL72, 36 Vera CPUs share a liquid-cooled rack with 72 Rubin GPUs. The CPUs coordinate the work, the GPUs perform the calculations, and a network called NVLink moves data between them. We look inside the rack to see why those connections matter. When an AI model has a long conversation to keep track of, moving information to the chip that needs it can be as important as the speed of the chip it... -
How Claude Helped Map a Vault Particle 06.11.2026 6mIn 1986, two researchers spotted a tiny structure whose curved ribs reminded them of a cathedral ceiling. They called it a vault. Forty years later, scientists are still working out what these tiny barrels do in our cells. We follow a team at Anthropic that used Claude to search for relatives of these hollow protein shells. One promising lead came from a virus that infects bacteria. The researchers grew its protein in E. coli and used an electron microscope to see how 96 copies fit together ... -
From Tokens to Bytes: AI Learns Its Own Building Blocks 05.11.2026 5mMost language models begin with a tokenizer that cuts text into predetermined pieces. What if the model could learn those pieces for itself? Cartesia’s hierarchical networks, or H-Nets, start with raw bytes and build useful chunks inside the model. We explore how this learned hierarchy can compress familiar patterns, preserve important detail, and even change its segmentation as new words and repeated structures appear within a passage. Cartesia reports promising efficiency gains in controll... -
Life on the Moon: Designing a Settlement Worth Living In 04.11.2026 5mWhat would it take to make the Moon a place to live, work, and build a community? The Aurelia Prize invites designers, architects, and engineers to answer that question with imaginative but technically grounded settlement concepts. We explore a competition offering a winning award of up to $100,000, and why its brief puts human dignity, scientific exploration, and everyday comfort alongside the engineering challenges of power, radiation, and lunar dust. The challenge reaches beyond a strikin... -
When AI Chooses: Jev, Confidence, and Better Decisions 03.11.2026 5mAn AI system often needs to make a small, specific choice: which team should handle a message, which category fits a document, or whether a request needs clarification. We explore Jev’s Choice primitive in TypeSafe, which returns a selected option together with probabilities across the available alternatives. Those extra signals make it possible for ordinary application code to handle ambiguous inputs more thoughtfully. A customer-support example shows why that matters. One message can conta...
Popular in
The podcast also appears in the podcast charts of these countries.