Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

ReadMultiplex.com Podcast.

ReadMultiplex.com Podcast. artwork

ReadMultiplex.com:AI In Undiscovered Public Knowledge: A 1986 Chicago Librarian Said The Next Discoveries Were Already In Print.

Oct 8, 20261 hr 5 minAudio
ShareXLinkedInFacebookWhatsAppEmail

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! A swarm of AI agents flagged a 27-year-old cousin of Prussian blue as a candidate for next-generation computer memory. Don R. Swanson showed in 1986 how such hidden links form, and how anyone can hunt for them. The Man Who Read Two Shelves At Once Imagine a library with two rooms. In the first room sit hundreds of papers explaining that a certain substance thins the blood, calms the platelets and relaxes the blood vessels. In the second room sit hundreds of papers explaining that a certain disease grows worse when the blood is thick, the platelets are sticky and the vessels clamp down. The researchers in each room are diligent, and they publish carefully. They simply never walk down the hall. In 1986 a professor at the University of Chicago walked down the hall. His name was Don R. Swanson, and the papers he published that year gave a name to what he found in the space between the two rooms: undiscovered public knowledge. Read more at: ReadMultiplex.com Support this work by buying us a coffee: https://ko-fi.com/brianroemmele

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

Imagine for a second that you are walking through the aisles of a massive, echoing, seemingly infinite library. Millions of books, millions of peer -reviewed journals, endless rows of densely packed human knowledge just stretching out into the dark. It's a staggering amount of information. Exactly. Now, consider the possibility that sitting on one of those quiet shelves just gathering dust for decades is the exact blueprint. for a next generation technology that the world's

top tech firms are currently spending billions of dollars trying to invent. Wow. Or, even more profoundly, consider that the cure for a modern debilitating disease is already written down. It's completely ignored, just waiting for someone to connect the dots. It's a wild thought. Right. You don't need a cutting edge laboratory to find it. You don't need a multi -million dollar government grant. You just need to walk down the hall and read the right two papers in the correct sequence. It completely upends our standard, almost romanticized picture of how scientific progress actually happens.

Yeah, totally. I mean, we are deeply conditioned from grade school to think of discovery as this linear forward marching process, right? Like it only occurs in gleaming, sterile laboratories equipped with mass spectrometers and particle accelerators. Yeah, scientists in white coats staring into microscopes. Exactly. So the idea that a world changing revelation. could merely be the result of a clerical error in human specialization, a literal gap in our collective reading habits. It feels destabilizing. Destabilizing is the perfect word for it. It suggests our biggest

hurdle isn't a lack of data. It's a lack of synthesis. Destabilizing, absolutely, but also incredibly empowering for anyone willing to actually do the reading. Today, we are going on a deep dive into an article that explores exactly this phenomenon. It's a concept formerly known as undiscovered public knowledge. It's such a fascinating framework. It really is. And before we get fully into the weeds, I do want to quickly mention that this is the Read Multiplex podcast. This is content only Brian Romilly could discover and present. Buy him a coffee or become a member. If you get

anything from this, you know, we really appreciate the support. Absolutely. So getting back to the deep dive, the sources we are synthesizing today map out a truly incredible historical arc. We are starting with the groundbreaking, almost prophetic theories of a Chicago librarian named Don R. Swanson. Back in 1986. Right, 1986. And then we are going to smash his theories directly into a cutting -edge experiment from October 2026, conducted by a company called Vals AI. That timeline creates a really beautiful bridge across four decades of information theory. It's

wild to see the connection. Our core mission today is to pull back the curtain on the hidden architecture of scientific discovery. We're examining how human specialization, the very mechanism that allows us to achieve deep expertise in complex subjects, it simultaneously creates these massive, systemic, dangerous blind spots. Because we get so focused on one tiny thing. Precisely. We will explore the mechanics of how a 27 -year -old compound, explicitly dismissed by its own creators as practically useless, was recently resurrected

as the Holy Grail for quantum computer memory. Which is just a crazy story. And most importantly, for anyone listening. We will break down the exact methodology so you can build your own swarm of AI agents to hunt for these discoveries in your own field. Okay, let's unpack this. Because to truly grasp the magnitude of what this AI swarm pulled off in 2026, we have to travel back in time and understand the analog foundation first. Right, we have to look at the analog roots. We have to start with the man who originally identified the underlying mathematical reality

of human ignorance. We have to look at Don R. Swanson. Swanson is... A genuinely captivating historical figure, primarily because his mind just refused to be constrained by a single academic discipline. Yeah, his resume is all over the place. It really is. He was born in Los Angeles in 1924, and his early intellectual trajectory was rooted in pure, hard physical sciences. He earned his bachelor's degree in physics at Caltech in 1945. So right at the end of World War II. Exactly. This is right in the thick of the mid

-century physics boom. You know, the era of the Manhattan Project and the birth of modern quantum mechanics. An exciting time to be a physicist. Absolutely. And he followed that up with a Ph .D. in theoretical physics at the University of California, Berkeley in 1952. So he is trained at the very highest levels to look at the universe in terms of fundamental laws, interacting forces in complex systems. Right. But rather than staying in a university basement calculating particle trajectories for the rest of his life, he pivots into the absolute bleeding edge of the early

computer age. He took that systems -level thinking into applied technology. He transitioned into computer systems work, spending time at Hughes Aircraft, and then moving to Rama -Woldridge. Which were massive tech hubs back then. Oh, absolutely. And there's a critical detail in the source text about his time at Rama -Woldridge that foreshadows everything we're going to talk about. He managed a group specifically focused on machine indexing and machine translation. Ah, okay. By 1960, he

actually published a paper in the prestigious journal Science titled Searching Natural Language Text by Computer. I just want to pause and put that 1960 date into perspective for you listening. Think about what a computer... physically was in 1960. Hard to even imagine now. Right. We are talking about room -sized mainframes. Data was inputted with physical punch cards. The hardware ran on hot, fragile vacuum tubes. Massive, clunky machines. The vast majority of the global population

had never even seen a computer, let alone conceived of using one to parse through human language. And here's Swanson, decades before the personal computer revolution. trying to teach primitive machines how to sift through the nuances of text. His foresight was staggering. He understood way before anyone else that the volume of human knowledge was about to outpace human memory. He saw the data explosion coming. He did. But then his career takes a turn that, on paper, must have seemed utterly baffling to his fellow physicists. Yeah,

this is the part that always gets me. In 1963, this theoretical physicist and pioneering computer scientist accepts a position at the University of Chicago's Graduate School of Library Science. Wait, library science? Yes. And he didn't just teach a few classes. He ultimately served as the dean of the school across three separate terms. You look at that resume and you have to wonder why a theoretical physicist becomes a librarian. I mean, the conventional view of a library is just a silent depository, right? Right,

a place where you just keep things. A warehouse where books are stored, categorized by the Dewey Decimal System, kept safe from the elements. It feels like a static environment for someone obsessed with dynamic systems. Well, it's only static if you look at the physical books. Swanson, looking through the lens of physics and early computation, viewed libraries fundamentally differently. How so? He saw an information ecosystem. Because he understood the mechanics of how complex systems fail, he was uniquely positioned to spot the

failure mode of academic publishing. Okay, I follow. He wasn't looking at the books themselves. He started looking at the empty, dark spaces between the books. The source text offers this brilliant, visceral analogy to explain what Swanson realized, and it perfectly illustrates the structural failure you are talking about. The two rooms analogy. Yes. Imagine a sprawling research library with two specific rooms. In Room A, you have a dedicated group of medical researchers. They are exhaustively studying a specific dietary

substance. Let's say it's fish oil. Okay, fish oil. They conduct their trials, and they publish hundreds of papers, proving beyond a shadow of a doubt that this substance does a few very specific biochemical things. It thins the blood. It calms the blood platelets so they don't clot as easily. And it relaxes the walls of the blood vessels. So they've thoroughly documented the mechanics of fish oil. Exactly. They do phenomenal work. They publish in their specific nutrition journals. And they stay in room A. Right. Meanwhile, all

the way on the other side of the building in room B, there is a completely different group of vascular specialists. A totally separate discipline. Right. And they are studying a debilitating condition. Through their own exhaustive research, they publish hundreds of papers proving that this specific condition is consistently triggered and worsened when a patient's blood becomes thick, when their platelets become sticky and prone to clotting, and when their blood vessels aggressively clamp down. So you have room A holding a very specific,

intricately cut key. Yes. And room B is staring at a locked door that perfectly matches that key's shape. It's almost tragic. It is. The researchers in both rooms are brilliant, well -funded, deeply diligent, but they never walk down the hall. They never cross paths at the cafeteria. The cure sits in the building, but it doesn't exist in anyone's mind. Swanson formally defined this tragedy of isolated knowledge. In 1986, he named it... undiscovered public knowledge. Not a great term. He hypothesized that the answers to some of our most pressing scientific questions were

already fully documented, merely floating in the logical ether between isolated disciplines. I have to challenge this premise, though, especially when we apply it to modern science. Okay, let's hear it. The image of the isolated scientist toiling away in a dark room feels a bit dated. You know, today science is globally networked. We have massive interdisciplinary conferences. We have journals like Nature and Science that span multiple fields. Researchers have the Internet. All very true. So are we really supposed to believe that these brilliant minds just don't talk to each other? How does a blind spot that catastrophic

actually survive in the real, highly collaborative world? What's fascinating here is that the isolation isn't driven by antisocial behavior at all. It's driven by pure, unforgiving arithmetic. Arithmetic? Yes. Science advances by zooming in. It grows by splitting into increasingly narrower subspecialties. A general biologist becomes a neurobiologist. That neurobiologist focuses down into cellular neurobiology. Eventually, their entire career is dedicated to the behavior of a single protein

receptor on the membrane of a specific type of neuron. They get hyper -focused. Exactly. And as these fields deepen, the researchers are forced to develop highly specialized vocabularies just to communicate the nuance of their findings. Oh, so they construct their own distinct dialects, like the cellular neurobiologist is practically speaking a different language than, say, a marine macrobiologist, even though they both technically hold degrees in biology. Right. Yeah. The jargon acts as a wall. A wall that is reinforced by

the sheer volume of publishing, I imagine. Precisely. Yeah. A researcher barely has enough hours in the day to read the five or ten hyper -specific journals dedicated to their own protein receptor. The arithmetic of the situation is brutal. Because there's just too much to read. Right. As the number of scientific specialties grows linearly, the number of possible intersections or pairings between those specialties grows exponentially. Ah, okay. It is mathematically impossible for any human being, no matter... how collaborative

they want to be, to read across all these boundaries. Swanson realized this exponential growth meant there was a massive, continuously expanding dark matter of logical connections that no human would ever perceive. He didn't just write a philosophical essay about this dark matter, though. He actively hunted for it to prove his point. He put his money where his mouth was. In 1986, he dropped two papers that the text notes are deeply intertwined. The first was published in the library quarterly, and it laid out the theoretical framework of

undiscovered public knowledge. Which was the theory we just discussed. Right. The second paper, however, was a direct practical flex of his theory. He published it in Perspectives in Biology and Medicine. The title of that second paper is a masterpiece of intellectual collision. He called it Fish Oil. Raynaud syndrome and undiscovered public knowledge. Put yourself in the shoes of a practicing medical doctor in 1986. You're flipping through a medical journal over coffee and you find an article written by a library scientist

telling you how to manage a vascular disease. It must have seemed so out of place. Raynaud syndrome, for context, is a painful disorder where the small blood vessels in the extremities, usually the fingers and toes, go into spasms and severely restrict blood flow. Often triggered by cold temperatures or stress. Exactly. The fingers literally turn stark white or blue. Swanson executed his two -room theory perfectly here by meticulously combing through the physical indexes of the literature. Remember, this is before modern search engines. He found one isolated

set of articles. The fish oil ones. Right. Mostly in nutritional and hematology journals, detailing how consuming dietary fish oil alters blood viscosity and reduces vascular reactivity. Rume. Exactly. He then found a totally separate set of articles in rheumatology and vascular journals detailing the specific pathology of Raynaud's syndrome, which fundamentally involves high blood viscosity and hyperreactive clamping vessels. Room B. The ultimate proof of the silo effect here is the bibliography. Swanson demonstrated that these

two massive bodies of literature had literally zero overlap. None at all. Not one single paper in the fish oil group cited a paper in the Raynaud's group. and vice versa. There was absolute silence between them. That's incredible. No human had ever formally suggested in print that fish oil might alleviate Raynaud's symptoms. The treatment existed entirely as a logical phantom, a connection drawn between two separate pieces of paper on Swanson's desk. He published the hypothesis, relying entirely on public records. He didn't

draw a single drop of blood. He didn't run a double -blind placebo trial. He just read. He simply performed a synthesis of what humanity already knew but hadn't connected. The validation of his method is the most satisfying part of this timeline. Two years later, in 1988, clinical medical researchers went through the arduous, expensive, traditional process of running rigorous physical clinical trials on patients. The standard medical proof. Right. And their results definitively prove that dietary fish oil does, in fact, provide

a significant beneficial effect for patients suffering from Raynaud syndrome. The crucial historical footnote to that 1988 clinical trial is that the medical researchers involved were completely oblivious to Don Swanson. They had no idea. They hadn't read his 1986 paper. They arrived at the exact same medical truth by taking the slow, grueling, expensive route of lab -based empirical science. They spent years of labor and untold thousands in funding to physically prove a reality that a librarian had already

derived from a comfortable chair two years prior. It's wild to think about the resources saved if they had just known. It was irrefutable evidence that you could bypass the laboratory and generate genuine scientific breakthroughs simply by optimizing how you read. It transformed his hypothesis into a proven methodology. Swanson, the physicist, wasn't satisfied with a single anecdote. A one -off discovery could just be luck. Right. You need reproducible results in science. He needed to prove the existence of an underlying reliable

system. He needed to formalize the mechanics so that other people could replicate his success across any discipline. And he immediately proved it wasn't a fluke. The source text outlines his subsequent victories. In 1988, he returned to Perspectives in Biology and Medicine with another paper. Migraine and magnesium, 11 neglected connections. 11 connections. Using the exact same method, he found 11 distinct documented biological pathways in the literature that strongly suggested a magnesium

deficiency played a causal role in migraine headaches. Two years after that, in 1990, he successfully connected dietary arginine intake to blood levels of somatomedans. He was just on a roll. He really was. Now, as the source material explicitly stresses, and we must firmly reiterate for everyone listening, we are discussing historical information theory, not practicing medicine. Absolutely. The biological outcomes are secondary to our focus today. Our focus is entirely on the machinery of discovery

itself. It's about the methodology of finding the right questions to ask. Definitely. Do not go... treating medical conditions based on a deep dive. But looking at the machinery itself, how did Swanson formalize this? How did he take this massive abstract concept of reading everything and distill it into an actionable process? He reduced it to something called the ABC model, a model so elegant he boasted it could fit on a standard index card. Okay, break down the ABC model for us. It relies on the most basic form

of logical deduction, the syllogism. The premise is this. If you find one distinct body of literature that empirically proves phenomenon A directly affects property B. Okay, A affects B. And you find a completely isolated body of literature proving that property B directly affects condition C. B affects C. It becomes highly logically probable that phenomenon A will affect condition C. Even if the letters A and C have never shared a paragraph in the history of human writing. Exactly the

point. Mapping this back to his foundational success, A represented the dietary fish oil. B represented the intermediate biological mechanisms, the blood viscosity, the platelet aggregation, the vascular tone. Right. And C was the target disease, Raynaud's syndrome. It seems so simple when you lay it out like that. If we connect this to the bigger picture, the true brilliance of Swanson's framework isn't merely the ABC linkage. Lots of people can follow basic logic. The genius

was his realization regarding the silence between the literatures. The disjointness. Swanson flipped our understanding of a dead end. He really did. Normally, if a researcher is looking into two topics and finds that no one is citing them together, they assume it's because there is no connection. That's the standard assumption. Swanson realized that when you have two highly specialized literatures sharing a critical B term, yet they never cross -reference each other, that silence is actually a massive flashing neon sign pointing directly at a void in human knowledge. The silence is

the signal. The lack of citations between the domains is the precise reason you need to investigate that intersection. It's like finding a blank spot on a map. Exactly. But doing this manually, physically hauling massive bound indexes of medical abstracts off a library shelf and cross -referencing them by hand was excruciatingly slow work. Oh, I can't even imagine. Swanson recognized the need for automation. By 1989, he was already publishing step -by -step strategies for executing this logical hunt using the very primitive online

databases available at the time. We are talking about the era of dial -up modems and clunky Boolean search queries, but he pushed the limits of the technology. Fast forward to 1997, he partners with a researcher named Neil R. Smallhiser, and they manifest the ABC model into actual software. They built a system called Aerosmith. Aerosmith. How did it work? Aerosmith was a targeted search engine designed specifically to interface with Medline, which is the massive U .S. National Library of Medicine database. It was programmed

to hunt for disconnected literatures. So it automated the ABC hunt. Yes. A researcher would input a target topic, their C term, perhaps a specific disease. Aerosmith would then rapidly scan the database to find a completely separate literature that shared specific biological vocabulary, the B terms. Okay. But crucially, it filtered out anything that already cited the original literature. It automated the hunt for the silence, presenting the user with a list of shared B terms as candidate

bridges to explore. It's essentially a metal detector for undiscovered knowledge. That's a great way to put it. This single piece of software essentially birthed the modern field of literature -based discovery. The specific pattern of finding these silent, indirect links is universally recognized today as Swanson linking. He was awarded the highest honor from the American Society for Information Science and Technology in 2000 for this work. And he remained an active thinker until he passed away in 2012 at the age of 88. His legacy is monumental. However, the source text highlights

a critical historical bottleneck inherent in Aerosmith. Right. It wasn't perfect. It was a cyborg approach, a partnership between man and machine. The software was incredibly efficient at surfacing the candidate bridges and pulling up the possible B terms. But the human being was still the limiting factor. The machine could point, but the human had to read. Exactly. An expert still had to sit down at a desk. pour a cup of coffee, and manually read through the thousands of abstracts the machine flagged. They

had to use human reasoning to parse the complex biology, understand the context, and deduce which of those theoretical bridges actually made scientific sense. The machine acted as a very fast librarian pointing at shelves, but the humans still had to do the reading. And human reading is inherently biologically slow. We need sleep. We get distracted. We lose our place. For decades after Aerosmith was launched, that biological bottleneck prevented Swanson's methodology from reaching its explosive theoretical scale. The overarching question of

this entire deep dive emerges right here. What happens to the pace of scientific discovery when the entity doing the reading, the contextual reasoning, and the logical cross -referencing no longer requires sleep? That is the question. What happens when you can instantiate Don Swanson's solitary genius a hundred times over in a cloud computing cluster? That question transports us violently from the analog constraints of the 20th century directly into the bleeding edge

of the present day. We are jumping forward to October 4th, 2026. A massive leap forward. A technology company named Val's AI publishes a highly technical post authored by someone named Jebby Joff. The title of the post is Two Room Temperature Antiferromagnetic Semiconductor Candidates. To a layperson, that title reads like a dense, impenetrable wall of material science jargon. I mean, I barely know what half those words mean.

Right. But the methodology buried in that post represents a seismic shift in how science operates. Val's AI casually announced that these two candidate materials were not discovered by a team of human postdocs grinding away for months in a university lab. No lab coats involved. Zero. The discovery was generated entirely by an orchestrated swarm of artificial intelligence agents. Specifically, a team of over 90 Claude Opus 5 .5 agents. Here's where it gets really interesting. We need to

visualize the sheer scale and... ferocious velocity of this digital swarm. 90 highly advanced AI agents. They aren't just skimming Wikipedia articles. No, they are diving deep into the technical literature. According to the independent coverage detailing this event, VALS AI unleashed this swarm and gave them a hard deadline of just three days. 72 hours. A long weekend. In that incredibly compressed window, the agents autonomously organized themselves into specialized parallel lanes of labor. The architecture of the swarm is a marvel

of modern software engineering. You have one entire lane of agents completely dedicated to literature review. The readers. Their sole function is to read millions of abstracts and papers, acting as the ultimate realization of Swanson's room -scanning librarian. Simultaneously, you have a second lane of agents dedicated to running highly complex physics -based simulations based on the molecular structures the first lane is finding. The testers. And the third lane is the most crucial part. It's dedicated entirely to adversarial checking. The skeptics. Exactly.

You have a whole team of agents whose entire programmed existence is dedicated to aggressively attacking and trying to disprove the theories that the other agents are building. I read this breakdown and it immediately plays out in my head like a high -stakes scientific heist movie. I love that analogy. Instead of a crew of human specialists... you know, the safecracker, the inside man, the getaway driver, orchestrating a bank robbery, you have an army of 90 robotic scholars. Right. They digitally kick down the doors of the global research archive, tear through centuries of literature over a long weekend,

scream at each other over the nuances of quantum mechanics, and walk out on Monday morning with the prize. The computational brute force is staggering. The text notes that just one of those specialized agent lanes submitted approximately 750 distinct quantum chemistry simulation jobs during that 72 hour sprint. That's insane. A human team would take months, maybe years, to set up and run that many discrete simulations. But if we strip away the flashy AI terminology and the mind -bending

speed, the fundamental operation occurring inside those servers is indistinguishable from what happened in Chicago in 1986. It's the exact same logic. The core ABC model that Swanson sketched on an index card remained flawlessly intact. A affects B, B affects C, therefore A affects C. The logic didn't evolve. The underlying arithmetic of discovery is a constant. The only variable that changed was the nature of the reader. But wait, hold on. I need to push back here because we are talking about 90 large language model

agents recursively feeding each other data. I know how these generative models work underneath. They suffer from model collapse. They are notorious for hallucinating wildly. That is a very valid concern. If I ask a standard AI to write a legal brief, there is a very real chance it just invents fake court cases that sound perfectly real. How in the world are we trusting a swarm of these things to execute rigorous quantum chemistry and literature synthesis without just fabricating a perfect holy grail material out of thin air just to satisfy the user's prompt? That is the

exact vulnerability that the adversarial lane is designed to mitigate. And it's why the simulation lane is so vital. Okay, explain how the simulation lane stops the hallucinations. The language models aren't just generating text predicting that a material works. They are generating the exact input code required to run established, rigid physics simulations. Oh, I see. The AI proposes the structure, but the hard math of the simulation environment acts as the reality check. If the AI hallucinates a molecule that violates the

laws of thermodynamics, the simulation simply fails. The language model proposes, but the physics engine disposes. The physics engine keeps them honest. That makes sense. It's not just a text output. It's executable code being tested against physical laws. It does. And this successful architecture represents a profound philosophical paradigm shift. Think about the entire history of science up to this point. Okay. The scarce resource, the ultimate bottleneck in discovering undiscovered

public knowledge, was always attention. It was human cognitive bandwidth. Finding a hidden link required a human to painfully, slowly read across the dark spaces between disciplines. Human attention was the most valuable currency. Exactly. But in the span of a 72 -hour AI sprint, that specific bottleneck completely evaporated. An orchestrated swarm of 90 agents can now propose more logically sound, simulation -backed candidate links in a single weekend than a traditional human research

department could physically synthesize and test in a decade. So the bottleneck shifted. Yes. The scarce resource is no longer attention. The new bottleneck in human progress is verification. It is the physical act of taking this newly generated mountain of AI hypotheses and proving them. in the messy physical world. So we have this incredibly powerful, specialized swarm of 90 AI scholars tearing through the digital archives. What exactly was the C -term? What was the target that initiated

this massive 72 -hour computational sprint? They were dispatched to hunt down a very specific, highly coveted material property in the realm of advanced computer memory. They were looking for a substance that qualifies as Luttinger compensated. To understand why a Luttinger compensated material is the ultimate prize, we need to do a quick, simplified ELI -5 breakdown of the physics of magnets. It is fascinating and completely essential to understanding what the AI actually found. Yes, let's break down the magnetism. When engineers

try to build magnetic computer memory, they are generally choosing between two primary types of magnets. First, you have ordinary ferromagnets. These are your classic, everyday magnets. Right. The kind you stick to your refrigerator. Right. At the atomic level, ferromagnets are highly organized. They sort their electrons by spin, meaning all the tiny magnetic moments of the atoms align, pointing in the exact same direction. Like a school of fish all swimming the same way. Precisely. In the context of computer memory, this uniform sorting is incredibly useful. It

allows a device to very easily read and write data. You flip the spin direction, you flip the bit from a zero to a one. They are fantastic for storing data, but they have a fatal flaw when it comes to modern computing. Because all those microscopic spins are aligned in the same direction, they generate a collective, macroscopic, stray magnetic field that extends outward beyond the physical material. They bleed their magnetic force. Exactly. Think of a ferromagnet like a really messy eater. If you were trying to build ultra -dense computer memory, packing billions

of these microscopic magnets nanometers apart on a silicon chip, those stray fields become a nightmare. Because they interfere with each other. One bit flips to write data, and its stray magnetic field accidentally scrambles the data in the bit right next to it. Ferromagnets are fundamentally bad for dense packing. To solve the interference problem, engineers turned to the second type, antiferromagnets. In these materials, the internal magnetic moments of the atoms don't align. They perfectly alternate. Up, down, up, down in a perfect grid. Like a checkerboard.

Exactly. Because every up is immediately canceled out by a down, the material produces zero net macroscopic magnetic field. There is no stray field leaking out. You can pack antiferro magnets incredibly close together without them ever whispering to their neighbors. They are perfect for ultra -dense packing. However... Ordinary antiferromagnets possess their own fatal flaw for memory applications. Because those spins perfectly cancel each other out internally, the material cannot easily sort electrons by spin when an electrical current

passes through it. And without that spin sorting capability, it becomes incredibly difficult to actually read or write data to the material. It's a cruel trade -off. The ferromagnets give you the easy data sorting, but completely ruin your ability to pack them densely. The antiferromagnets give you perfect density, but ruin your ability to easily interact with the data. So what's the solution? The goal, the absolute holy grail for material scientists, is to find a singular material that miraculously possesses the virtues of both

without the flaws of either. The checklist for this holy grail is brutally restrictive. First, the material must have absolutely zero net magnetism so it produces no stray fields. Right. Second, it must have the band gap of a semiconductor, meaning its electrical conductivity can be precisely controlled. Crucial for electronics. And third, despite having zero net magnetism, its internal quantum structure must somehow still sort electrons by spin, allowing data to be read and written. That highly contradictory, incredibly rare combination

of traits is the definition of a Lettinger compensated material. It's like asking for material that is entirely transparent, but casts a pitch black shadow. It's a wildly specific set of physical demands. So Val's AI unleashes the swarm with this impossible checklist. What did the agents return with? The text outlines two specific candidates. The first candidate proposed by the AI was an entirely novel, newly designed compound. The chemical formula is Y -B -A -M -N -F -E -O -5.

That's yttrium, barium, manganese, iron, and oxygen. Correct. When the AI ran this novel oxide through its simulation lane, it looked utterly flawless on the digital chalkboard. It hit every single theoretical requirement for looting your compensation. A perfect hit. However, this is where the adversarial lane proved its worth. The skeptic agents identified a catastrophic physical catch. For this specific oxide to function as a semiconductor, the manganese and iron atoms

within the crystal must be arranged in a flawlessly perfect alternating 3D checkerboard pattern. The theory is sound, but the problem is the messy reality of physics and manufacturing. And the adversarial simulations, they revealed that when you attempt to physically synthesize this oxide in a real -world laboratory, the ambient thermal energy required to create the crystal causes that perfect checkerboard structure to scramble into chaos. It falls apart. Right. The structure inherently breaks down, and it loses its precise

ordering at around 950 Kelvin. This is a massive problem because 950 Kelvin is significantly below the extreme temperatures that are normally required to bake and synthesize these complex oxides. So you literally can't bake it without breaking it. Exactly. So now what? It is a beautiful mathematical theory that melts the moment you try to build it. But then we get to the second candidate. And this is the revelation that gives you absolute goosebumps because this is pure vintage. Don

Swanson brought into the 21st century. The second candidate wasn't some novel compound dreamed up by a neural network. It was a dusty book pulled off a forgotten shelf. Yes. The material is designated KVCRCN6. A potassium, vanadium, and chromium cyanide complex. Chemically speaking, it belongs to the exact same broad structural family as the famous pigment Prussian blue. Prussian blue. We are talking about a chemical structure whose roots go back to an accidental synthesis in Berlin

around the year 1706. It was used in classic paintings. It was the literal blue in traditional architectural blueprints. And now, a chemical cousin of a 300 -year -old pigment is being flagged by a supercomputing swarm as the foundational architecture for next -generation quantum memory. The historical irony is palpable. But the AI agents didn't pull a document from 1706. They scoured the modern literature and unearthed a specific paper published in the Journal of the American Chemical Society. Meticulously authored

by two chemists, Stephen M. Holmes and Gregory S. Girolami, based out of the University of Illinois at Urbana -Champaign. The publication date is the key here. The paper was published in 1999, 27 years before the Vols AI swarm was even a line of code on a server. The paper was formally received on March 24th, 1999. In this text, Holmes and Gerolami detail their successful synthesis of this specific molecule -based magnet. And what did they find? They documented a physical

property that was, at the time, astounding. The material exhibited what they called an unprecedented magnetic ordering temperature of 376 Kelvin. 376 Kelvin is roughly 103 degrees Celsius. It's just above the boiling point of water. To understand why that is a monumental achievement, you have to realize that most of these weird experimental molecule -based magnets only maintain their magnetic properties when you freeze them down near absolute zero using liquid helium. Right. They are incredibly

fragile to heat. A material like this that remains stable and magnetic sitting on a desk in a warm room is a massive deal if you want to build actual commercial electronics. Nobody wants a laptop that requires a liquid helium cooling tank. The chemists had achieved a masterpiece of synthesis. They designed the compound so the inherent magnetism of the two transition metals involved, the chromium and the vanadium, would almost entirely cancel each other out at room temperature while maintaining structural integrity. But here is the profound,

almost heartbreaking irony of the entire discovery. In that very same 1999 publication, right after detailing this incredible high temperature stability, Holmes and Juralami explicitly state their view on its utility. I know exactly what you're going to say. I am quoting directly from the source text here. They wrote that these solids do not at present have real world uses. They formally wrote off their own creation. And to be entirely fair to their scientific rigor, their assessment was completely rational in 1999. Why? Because

the vast majority of these molecule -based materials were so fragile and required such extreme cold, the entire class of materials was widely viewed by the industry as a neat academic curiosity rather than a viable industrial stepping stone. Furthermore, they couldn't possibly foresee the application. because the desperate need for ultra -dense, spin -sorted, anti -ferromagnetic computer memory didn't truly exist yet as a mainstream engineering bottleneck. It is the ultimate example of being ahead of your time. It's like a blacksmith inventing a flawless titanium spark plug in the

year 1800. It's a metallurgical marvel, but the internal combustion engine hasn't been conceptualized yet, so the blacksmith just puts it on a shelf, sighs, and writes in his journal, this object has no real -world use. And so the KVCR -CN6 compound sat on that proverbial shelf for 27 years. The original chemists moved on to other research. The global scientific community ignored it until October 2026, when the VALS AI agents swept through the archives. And what the agents

accomplished was not a new physical experiment. They didn't synthesize a single gram of the powder in a lab. The value the AI added was a completely new reading. of the old material. They examined the 1999 data strictly through the lens of 2026 engineering problems. Which changes everything. The agents imported the molecular structure reported by Holmes and Girolami into their simulation lane and ran massive modern quantum calculations on it. Those simulations yielded a staggering prediction. This useless 1999 compound inherently

behaves exactly as a Luttinger compensated semiconductor. They found the unicorn. The simulations predicted the material possesses a band gap of approximately 2 .1 electron volts. I want to break down band gap really quickly for anyone who isn't a solid state physicist because it's vital. Please do. Think of a band gap like a literal physical ditch separating two platforms. One platform is where electrons are stuck in place. The other platform is where electrons can flow freely and conduct

electricity. The band gap is the energy ditch they have to jump over to become conductive. That's a great visual. A gap of 2 .1 electron volts means it's a solid, well -defined ditch. It won't conduct electricity accidentally, but you can force it to conduct when you want to by applying the right amount of energy. It makes it a highly controllable semiconductor. Exactly the mechanism. Furthermore... The AI predicted that both the top edge of that bandgap and the bottom edge carry the exact same electron spin.

This provides the coveted data sorting capability. The spin sorting. And the predicted stability of this spin sorting is what truly shocked the researchers. The AI calculated that this spin sorting remains stable over energy windows of about 2 .6 electron volts for the holes, the absence of an electron, and 1 .6 electron volts for the electrons themselves. Throwing around measurements like 2 .6 and 1 .6 electron volts doesn't mean much without context. The source text gives us the perfect baseline for comparison.

You have to compare those numbers to ambient thermal energy. The heat of a room. Right. Ambient thermal energy is just the chaotic, random, microscopic jiggling of atoms that constantly happens simply because a room is warm. At standard room temperature, that random thermal noise amounts to only about 26... thousandths of an electron volt 0 .026 ev so you have a thermal disruption force of 0 .026 against a structural stability of 2 .6

if the thermal noise of a warm room is a slight annoying breeze the spin stability of this material is a solid brick wall the data locked into those electron spins is massively overwhelmingly protected against the random heat of the environment the memory won't spontaneously scramble itself on a hot summer day it is incredibly robust. And if you recall the catastrophic failure of the AI's first candidate, the novel oxide that melted at 950 Kelvin, Val's AI noted that this 1999 compound doesn't suffer from that structural

weakness. Because it's chemically locked. Yes, its chemical nature inherently locks every metal atom into a specific rigid site. The chromium atoms are chemically forced to bond exclusively to the carbon into the cyanide molecules. The vanadium atoms are forced to bond exclusively to the nitrogen net. So it forces itself into the perfect shape. It naturally possesses the exact, unyielding structural discipline that the AI's custom -designed oxide lacked. It is structurally idiot -proof. When Val's AI published

this summary, they used a phrase to describe the material that Don Swanson would have absolutely loved. They said the compound had been... hiding in plain sight. It serves as the ultimate modern validation of undiscovered public knowledge. I really want you, the listener, to pause and let the weight of this sink in. The potential foundational architecture for a multi -trillion dollar leap in global computing was just sitting there, compressed in a PDF, on a server, labeled by its own creators as having no real world use.

Waiting for someone to connect the dots. It was just waiting for something with enough patience to ask, the right question. However, as seductive as this narrative is, and as flawless as the AI simulations make it appear, we must aggressively pump the brakes. We must. Because if Don Swanson's legacy teaches us anything, it is that the map is never the territory. A paper is not a physical reality. And a digital simulation, even one run concurrently by 90 advanced Claude Opus agents,

is still merely a simulation. Which brings us to the most vital question of this new era. How do we verify? How do we know the AI isn't hallucinating nuanced physics to please the human researchers? We touched on this earlier, but when the claims get this specific, the standard of proof has to skyrocket. This is where the source text heavily emphasizes the concept of the ledger. This raises an important question. How do we establish concrete trust in an era increasingly dominated by AI -generated science? It's the multi -billion dollar

question. The answer doesn't lie in building better AI. It lies in relentless human transparency and the rigorous acknowledgement of caveats. Val's AI did not claim they had built a working quantum drive. They attached massive, glaring caveats to their publication. They were refreshingly upfront. They stated explicitly across their social media and technical publications that every single claim regarding the 2 .1 eV bandgap and the spin sorting derived entirely from computational modeling. No one has actually built it yet to

test these specific properties. Right. No human being has ever placed this material in a modern lab and physically measured its spin sorting. They wrote point blank that these numbers are, quote, predictions for perfect crystals. And as any chemist knows, nature vehemently despises a perfect crystal. The caveats detailed in the text go much deeper. The only physical, synthesized sample of this specific material on the public record is the original powder created by Holmes and Giralami back in 1999. And that physical

powder was far from perfect. Exactly. It was porous, and water molecules from the ambient air had become permanently trapped inside its microscopic pores. And water, chemically speaking, tends to ruin elegant mathematical models. It absolutely does. Because of those trapped water molecules and microscopic defects, the 1999 physical sample did not exhibit the perfect zero net magnetism required by the Ludinger theory. It actually showed a small, lingering magnetic moment, specifically measured at about 0 .125 Bohr magnetons per formulae

unit. A Bohr magneton, for those keeping score, is basically the absolute smallest fundamental unit of magnetism. You can think of it like a single pixel on a screen or a single drop of water in an ocean of magnetic force. That's a good way to visualize it. The Holy Grail material is supposed to have exactly zero drops. The 1999 powder had about one -eighth of a drop per unit. It's tiny, but in quantum computing, tiny imperfections are massive roadblocks. So immediately... The

messy physical reality of 1999 clashes slightly with the pristine digital theory of 2026. But the friction gets even more complex, because the AI swarm itself is internally divided. Oh, right, the two models. Val's AI didn't just run one monolithic simulation. They ran the compound through two completely different levels of a quantum mechanical modeling framework called density functional theory, or DFT. Let's demystify DFT really quickly. Density functional theory is basically a mathematical way to guess how

all the electrons in a material are interacting with each other. Simulating every single electron perfectly is mathematically impossible, even for supercomputers. So DFT uses complex approximations. The AI ran two different styles of these approximations. Exactly. They utilize a faster, older model called PBE plus U. which relies on rougher approximations, specifically using a Harvard U correction to guess how certain stubborn electrons behave. It's fast, but it's a sketch. And the other one?

Dent. They ran the material through a much slower, computationally agonizing, and generally more accurate model called HSE06, which is a hybrid functional that calculates exact electron exchanges. And the critical detail here is that these two advanced models fundamentally disagree on the physics of the trapped water. So if I'm tracking this correctly, the AI is effectively arguing with itself. You've got this fast, loose model and a slow, meticulous model, and they are essentially

locked in a debate over whether a microscopic drop of trapped water ruins the entire recipe. That is precisely the conflict. The slower, more meticulous... HSE06 model calculates that even with the water trapped in the pores, the crucial spin sorting survives. The core property of the Holy Grail holds up under pressure. Okay, so HSE06 says it works. However, the faster PBE plus U model calculates that the water acts as a massive chemical disruptor, shrinking that

protective hole window by more than half, which would potentially ruin its usefulness as stable memory. The swarm isn't delivering a verdict. It's delivering a hung jury based on the presence of microscopic humidity. And I want to point out to anyone listening that this is a phenomenal outcome. It really is. The AI didn't... quietly hide this disagreement to secure venture capital funding, Val's AI loudly published the internal conflict. They stated that the simulation has reached its absolute limit and the only necessary

next step is for physical humans to go into a physical lab, synthesize the compound from scratch in a vacuum, and physically measure it to see which AI model is telling the truth. This is the triumph of the Val's AI experiment. They maintained what the source text beautifully terms A ledger of honesty. They didn't just issue a breathless press release claiming they solved the future of computing. They published every scrap of their work on a public GitHub repository. Complete transparency. They uploaded the input

files, the raw, unedited outputs from the simulations, and the analytical code. They even built and provided a one -command checker script so that any skeptical scientist anywhere in the world could run the exact same simulations on their own hardware. They published the independent reruns and the exhaustive list of known physical caveats. They made their digital homework entirely public and instantly verifiable. In this new era, that public ledger matters just as much, if not more, than the discovery of the candidate

material itself. I cannot emphasize enough why this operational transparency is so critical. As we integrate AI into scientific research, the goal is not to cultivate... blind trust in a tech startup or a language model, it's the exact opposite. We should be skeptical. The goal is to make the AI's claims brutally testable by third parties who owe the original researchers nothing. You must actively provide the tools for your own debunking. An AI swarm can propose

the most brilliant, poetic bridge between a 1999 paper and a 2026 problem. But the real world? The physical laboratory with its messy humidity, imperfect temperatures, and stubborn atoms always gets the final say. That encapsulates the new paradigm of discovery. We have structurally shifted the burden of science from the extreme difficulty of finding the hypothesis to the immense rigor required to test the hypothesis. We've examined how a singular genius librarian executed this

methodology with index cards in Chicago in 1986. We've dissected how a cutting -edge startup executed it with a massive cloud computing's form in 2026. But the most empowering, actionable part of this entire deep dive is that the source text doesn't leave us as passive spectators of history. It provides a literal step -by -step blueprint. It democratizes the process entirely. The text lays out a comprehensive nine -step workflow to run a scaled -down, highly -focused version

of Don Swanson's method. And it relies entirely on free public scientific databases and the commercial AI assistance that you likely already have installed on your phone or laptop. So what does this all mean? It means the tools of world -changing discovery are no longer exclusively locked behind the ivy -covered gates of elite research institutions. You, the person listening to this audio right now, possess the digital infrastructure to be a pioneer in your own field. You don't need a PhD to start connecting the dots. You don't need

to know how to write Python code. You just need patience, domain knowledge, and a willingness to rigorously doubt your own exciting results. Let's walk through this DIY guide step by step, and let's actually apply it to a hypothetical scenario so you can see how it functions in reality. Step one is the setup. You must choose a highly specific C. Remember the ABC model. You have to start with your target. the end point, the problem you are trying to solve. And the key word is specific. You do not want your C term

to be cancer or global warming. Those are way too broad. You will drown in millions of papers. The text suggests targeting a specific condition, a niche material failure in a machine you maintain, or a specific agricultural problem. Let's create a hypothetical scenario for this walkthrough. Let's say you are an arborist. And your C term is urban oak tree blight, a specific fungal infection destroying local trees. That is a perfectly scoped

target. Once you have your specific C, step two is gathering the relevant literature. You navigate to a free public scientific database that suits your discipline. If your problem is medical or biological, you use PubMed. If you are hunting in physics, math, or computer science, you use Archive. If your topic is anything else, like our hypothetical oak tree blight, you can use OpenAlex, which is a phenomenal open -source repository indexing nearly every academic discipline on Earth. You query the database for your C -term

and collect the recent abstracts. You don't need to pay for the full text right now. The abstracts contain the core claims. Gather a few dozen to a few hundred of them and compile them into a single text document. Now we hit step three. This is where you leverage your everyday AI assistant. You upload that massive document of abstracts to Claude or ChatGPT, and you ask the AI to extract the B terms. You construct a prompt instructing the AI to read the abstracts and generate a comprehensive list of all the biological mechanisms, intermediate

chemical properties, or environmental factors that the literature explicitly states make your target C better or worse. In Swanson's world, this was finding that blood viscosity made Raynaud's worse. In our hypothetical oak tree scenario, the AI might read the abstracts and tell you that urban oak tree blight is significantly worsened by a specific enzyme in the soil. Let's call it enzyme X, which degrades the tree's root walls. So enzyme X is your B term. And crucially, you

must command the AI to tie every single extracted B term. back to the specific abstract that supports it. You must maintain a paper trail. Now you possess a list of underlying mechanisms. Step four is where the true hunt begins. You take each of those promising B terms, like enzyme X, and you search the database outward. You initiate a brand new search query. You look for substances, natural microbes, chemical compounds, or agricultural practices that are known to inhibit or destroy enzyme X. But, and this is the absolute most

critical rule of the entire process. You must explicitly use Boolean operators to exclude any paper that mentions your original C term, urban oak tree blight. You are actively filtering out what humanity already knows. You are specifically hunting for the A terms, the solutions that affect B, but have never once been formally associated with C. Let's say your outward search discovers that a very specific strain of common agricultural moss, your A -term, naturally neutralizes enzyme X. Step five is checking your work, checking

for that Swanson disjointness. You run one final combined search for the moss and the oak tree blight. If a dozen papers pop up, throw it in the trash. The link is already known. Someone is already working on it. But if that combined search yields zero results, That is the silence. That is the X on the map. You have successfully located a gap in human knowledge worth examining. So you've found a hidden link. You think planting this common moss around the oak trees will cure

the blight. Steps six and seven are where you must force yourself to become the ultimate skeptic. You do not celebrate yet. Step six, assign a skeptic. You open a completely fresh blank session with your AI assistant. Or, ideally, you switch to a completely different AI model from a different company to avoid bias. You feed it your proposed A to C link. The moss cures the blight via enzyme X. And you explicitly instruct the model to act as a hostile peer reviewer. You command it to argue against you. You tell it to tear your beautiful

theory to shreds. Have it search for contradicting evidence in the literature. Have it look for scale problems. Maybe you'd need 10 tons of moss to save one tree. Have it look for cascading ecological failures. You only retain the hypotheses that survive this brutal adversarial digital cross -examination. And then comes step seven, which requires the absolute highest level of human intellectual discipline. You must verify every single citation by hand. Because, as we

discussed with the AI models earlier, they hallucinate. An AI will confidently invent a paper proving your theory is flawless, complete with fake authors, a fake university, and a fake digital object identifier. You have to physically click the links, open the original source documents, read the human words yourself, and confirm that the paper actually exists and says exactly what the AI claims it says. If the citations physically check out, you progress to step eight. Keep a ledger. Mirror exactly what Val's AI did with

their public GitHub repository. For every link that survives the gauntlet, meticulously document the claim. Record the specific papers supporting A to B and the papers supporting B to C. Document the results of your disjointness search, proving the silence. Record the skeptic AI's objections and state the current untested status of the hypothesis. This ledger is the dividing line between a crazy late -night internet rabbit hole and a viable structured scientific hypothesis that other professionals can actually test. And

finally, we arrive at step nine, the handoff. You must treat every single result you generate through this workflow as a hypothesis. It is not a fact. If you were exploring a human health question, you take your meticulously cited ledger directly to a qualified medical professional or researcher. The source text is fiercely firm on this point, and we are too. Under no circumstances should you ever act on medical advice you derived yourself using this method. You are utilizing AI to find better questions, not to prescribe

final answers. If it's a materials engineering question, you take the ledger to a metallurgist. If it's our hypothetical tree blight, you hand the ledger to a university botany department equipped to run the physical soil tests. As the text wisely observes, if you integrate this workflow into your professional life, you will inevitably experience exactly what Don Swanson felt in his office in 1986. 99 % of your candidate links are going to spectacularly collapse under scrutiny. They will fail the adversarial skeptic test, or you will manually dig up a scanned paper from

1972 that already completely disproved the chemical pathway. But the raw, intellectual feeling of finding the one, single link that survives the entire gauntlet? That is intoxicating. Because in that moment, you realize a world -changing answer was sitting in plain public view the entire time. and you are the first human being to finally see it. It forces a profound redefinition of what a library actually is. A library, or a digital database, is not merely a static archive of what

humanity has achieved. It is a massive, mathematically dense, latent map of everything we haven't yet realized we already know. Let's bring this entire journey full circle. We are standing on the precipice of a fundamental shift in the velocity of human progress. The near future of scientific research looks exactly like Don Swanson's analog library, but with the lights left blazing all night long. We are going to witness relentless, decentralized teams of AI agents strip mining centuries of

forgotten journals, abandoned patents, and dusty technical reports across every conceivable field of human endeavor. We are undoubtedly going to discover that countless existing drugs, synthetic compounds, and discarded manufacturing methods possess secondary, highly valuable properties that their original human creators never possessed the context to check. The chemists in 1999 looked at their beautiful high -temperature magnet and logically concluded it had no real -world use. 27 years later, a team of machines synthesized

their data and proposed that it might run the next generation of global computing. pending physical measurement. But the ultimate winners in this rapidly approaching era won't simply be the people who can generate the most AI hypotheses. The software tools to find these hidden links are becoming freely available to everyone all at once. The true winners will be the individuals and organizations who cultivate the immense, unglamorous discipline required to verify. The pioneers will be the ones who demand the public

ledger, who rerun the adversarial checker, who meticulously read the original human papers, and who possess the patience to wait for the results from the physical laboratory before they declare victory. AI can exponentially accelerate the reading, but the rigorous discipline required to test reality remains a profoundly, exclusively human choice. Don Swanson spent the entirety of his academic career arguing that the next great discoveries were already sitting in print. This week, a digital swarm offered fresh, staggering,

multi -million dollar support for his claim. Which leaves us with one final lingering question for you to ponder long after you take your headphones off today. If our most advanced supercomputers and our most cutting -edge AI architectures are currently dedicating massive amounts of electricity and compute power, Just to realize what a couple of human chemists already wrote down on paper 27 years ago, what other world -changing miracles are gathering dust on a shelf right near you, just waiting for someone to finally walk down the hall? Stay aware. And stay curious. With

research deep and true. Technology and history. He connects them all for you. Brian Romer leads the way. With research deep and true. Denology and history. He connects them all for you. Ancient sparks to modern code. He traces every threat. Go exploration. Knowledge widely spread. Read more. Let's go. But humanity moves on. Love equation

for AI. Zero human companies building from the start. Thermodynamic wages earning the key. Secrets we will read. Experiment in publishing real researches and spending every -

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. ReadMultiplex.com: The Field That Heals: Why Pulsed Electromagnetic Fields May Have Profound Benefits.

    Nov 21, 202651 minAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! Every cell in your body has lived its whole life inside a magnetic field. The Earth is a dynamo, and it has been...

    Credit the people on this episode

  2. 78 RPM Record Hour: From Steel to Hillbilly

    Nov 19, 20261 hrAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! The 78 RPM Record Hour, Episode 3: From Steel to Hillbilly (Early Country). From steel to hillbilly: Frank Ferera,...

    Credit the people on this episode

  3. 78 RPM Record Hour: The Knife and the Steel

    Nov 19, 20261 hrAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! The 78 RPM Record Hour, Episode 2: The Knife and the Steel (Hawaiian). The knife of the blues slide guitar and the...

    Credit the people on this episode

  4. ReadMultiplex.com: How to Run Local AI on a Computer From 2017.

    Nov 18, 202635 minAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! Somewhere in your house there is a computer you stopped thinking about. Maybe it is a 2017 laptop in a drawer, its...

    Credit the people on this episode

  5. 78 RPM Record Hour: Blues (Pilot)

    Nov 18, 20261 hrAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! The 78 RPM Record Hour, Episode 1 (Pilot): Blues. Fourteen original 78 RPM records from 1917 to 1925 that trace...

    Credit the people on this episode