Category: History

Categories
Business History IBM Infrastructure Nvidia Programming Semiconductors

The Half-Life of Moats

Prompted by an article on X by @magicsilicon on the CUDA moat. Research and drafting assistance from my AI intern assistant Clark.

The NVIDIA H100 looks, in retrospect, like an inevitability. It wasn’t.

What Jensen Huang built is more accurately understood as a sixteen-year accumulation of optionality — a platform investment made in 2006 for a market that wouldn’t fully materialize until 2022. NVIDIA intros the G80 architecture in November 2006, laying the groundwork for CUDA’s release a few months later. The stated ambition was to let scientists write C++ that ran on GPU cores without needing to understand 3D graphics pipelines. The unstated bet was that parallel computation would eventually matter for something bigger than rendering shadows in video games.

For sixteen years, it mostly didn’t. Not at scale. Not commercially. CUDA lived in research labs and HPC clusters. It attracted a small, devoted, and economically marginal user base — the kind that papers cite but investors ignore. NVIDIA kept investing in it anyway: cuDNN for deep learning operations, cuBLAS for linear algebra, a layered ecosystem of libraries that made CUDA not just accessible but nearly irreplaceable for anyone doing serious numerical computation. When TensorFlow and PyTorch emerged as the standard frameworks for neural network research, they didn’t adopt CUDA because it was the only option. They adopted it because CUDA was where the optimized kernels already lived.

AlexNet won the ImageNet competition in 2012 and did it on two NVIDIA GPUs. The deep learning community noticed immediately. The financial community largely did not.

Then ChatGPT launched in November 2022, and suddenly everyone needed H100s they couldn’t get.

The parallel to Intel is instructive and also undersells how strange this kind of story looks while you’re living through it. Intel was founded in 1968 as a memory company. DRAM. The founders — Noyce, Moore, Grove — were materials scientists and engineers who believed the future was in silicon memory chips. They were right, briefly: in the early 1970s Intel dominated the DRAM market. By 1984, that share had collapsed to 1.3%, ceded almost entirely to Japanese manufacturers who had commoditized the product.

What saved Intel wasn’t a pivot so much as a realization that a stopgap had become a foundation. The 8086, conceived in 1976 as an internal hedge and launched in 1978 was never supposed to matter. It was a 16-bit processor designed to hold off Zilog while Intel finished its ambitious 32-bit iAPX 432 architecture. The 8086 was assigned to a single engineer. “If management had any inkling that this architecture would live on through many generations,” its designer Stephen Morse later recalled, “they never would have trusted this task to a single person.”

IBM chose the 8088 — a cost-reduced variant — for the original IBM PC in 1981. That decision wasn’t destiny, it was simply a procurement. And yet from that accident of selection, Intel’s x86 line became the backbone of personal computing for four decades. The Pentium in 1993 was Intel’s Wintel moment — the flag bearer the @magicsilicon tweet gestures at — but the flag had been quietly sewn since 1978.

What these histories share is not just a pattern of “slow build, explosive payoff.” The structural similarity is subtler: in both cases, the moat was a software abstraction layer built on top of hardware. Intel’s real lock-in wasn’t transistor count or clock speed. It was backward compatibility — the commitment, formalized with the 80386 in 1985, that every future Intel chip would run software written for older ones. That promise created a flywheel that trapped developers and buyers in a virtuous (for Intel) dependency loop for decades.

CUDA is the same architecture at a different layer. The lock-in isn’t the H100’s 80 gigabytes of HBM3. It’s that switching to an AMD MI300X or Google TPU means potentially rewriting training pipelines that have been optimized against CUDA kernels for years. AMD’s ROCm platform exists. It is, by most accounts, maturing. Engineers who have tried the migration report that it costs months and hundreds of thousands of dollars. The moat isn’t a wall. It’s accumulated friction — the switching cost of a decade of engineering decisions baked into codebases that no one wants to touch.

But to find the actual origin of this pattern, you have to go back further than Intel. To 1964, and to a decision IBM made that Fred Brooks — its project manager — called a bet-the-business move.

The IBM System/360 was announced on April 7, 1964, after five years of turbulent internal development. What it introduced wasn’t just a new computer. It was a new concept: the separation of architecture from implementation. Before the 360, IBM ran five incompatible product lines simultaneously. A customer who outgrew their machine had to scrap all existing software and start over. The 360 replaced all five lines with a single unified architecture — six models covering a fiftyfold performance range, all running the same operating system, all sharing the same instruction set. The name itself encoded the ambition: 360 degrees, all directions, all users.

Gene Amdahl, the 360’s chief architect, had a precise formulation for what this meant: the architecture was “an interface for which software is written, independent of any implementation.” The Principles of Operation manual described what the machine did; separate Functional Characteristics documents described how each model did it. This distinction — separating the contract from the execution — was genuinely new. It’s the conceptual root of everything that came after.

The 360 generated over $100 billion in revenue for IBM and established the first platform business model in computing. Jim Collins would later rank it alongside the Model T and the Boeing 707 as one of the three greatest business achievements of the twentieth century. But its deepest legacy was architectural: the insight that if you make your abstraction layer the standard, the hardware underneath becomes fungible. Customers didn’t buy specific IBM machines. They bought into OS/360. The machines were an implementation detail.

Intel understood this by the 1980s, even if implicitly. The 80386’s backward compatibility commitment in 1985 was IBM’s 360 insight applied to microprocessors — the architecture is the product, the silicon is the vehicle. CUDA is the same insight applied to GPU compute. What NVIDIA sold researchers in 2006 wasn’t the G80 card. It was the abstraction: write parallel code in C++, run it on any NVIDIA hardware, trust that the next generation will be faster and compatible.

The pattern is now sixty years old. It has reproduced in every major platform transition. And it keeps working for the same reason it worked in 1964: when you own the layer that developers write to, your customers’ switching costs compound every year they stay.

There’s something worth sitting with here. Neither Jensen Huang in 2006 nor Gordon Moore in 1968 could have specified exactly what the payoff would look like. What they shared was a willingness to build infrastructure for a demand they could sense but not yet see — and the discipline to keep investing in it through the long years when it looked like a research project rather than a business.

The question that doesn’t resolve cleanly is whether that kind of patience is a strategy or a personality. And whether, in an industry that now moves faster than the cycles it’s lived through, sixteen-year moats are still the kind that get built.

Which raises the uncomfortable corollary: the same AI tools that CUDA enabled may be what ultimately erodes it.

The attack on CUDA’s moat is now structurally different from anything AMD or Intel could mount before. OpenAI’s Triton compiler lets developers write GPU kernels in Python without touching CUDA at all, and generates optimized machine code that often matches hand-tuned CUDA performance. MLIR — Multi-Level Intermediate Representation, originally from Google — provides a compiler infrastructure that can target any hardware backend from a single codebase. AMD’s ROCm has historically been dismissed as immature; ROCm 7, released this year, delivers meaningfully better inference performance than its predecessors. And perhaps most directly: Claude Code reportedly ported a CUDA codebase to AMD’s ROCm in thirty minutes — work that previously took months of engineering time.

The irony is almost too neat. CUDA’s moat was built on accumulated switching costs: the friction of rewriting code, the library dependencies, the tribal knowledge encoded in a decade of kernel optimizations. AI coding tools are specifically good at exactly that kind of mechanical, high-context translation. The weapon is attacking the wall it was built behind.

That said, it’s worth being careful about the speed of this. Abstraction layers that “should” erode moats often take far longer than expected, because the moat isn’t just the code — it’s the ecosystem of tooling, documentation, community knowledge, and hardware-software co-optimization that took eighteen years to compound. Triton and MLIR are real. They’re also early. The question isn’t whether the moat is vulnerable; it’s whether it erodes before NVIDIA’s next generation of chips makes it irrelevant to argue about.

As for what comes next — which company is building the IBM 360 of this decade — the honest answer is that it’s too early to call with confidence. But there’s a candidate worth watching.

Anthropic’s Model Context Protocol, launched in late 2024, has the structural fingerprint of a platform play. MCP is a standard for how AI agents connect to external tools and data sources — a common interface layer, hardware-agnostic (or rather, model-agnostic), that any system can implement. By late 2025 it had been donated to the Linux Foundation, adopted by OpenAI and Google, and was tracking 97 million monthly SDK downloads. There are now over 10,000 MCP servers. It is becoming the way agents talk to the world.

The parallel to OS/360 is imprecise but instructive. What IBM built in 1964 was a standard interface between software and hardware that decoupled what you wrote from what you ran it on. MCP is attempting something similar one abstraction layer higher: decoupling what an agent does from the specific models, tools, and data sources it does it with. If it becomes the standard — the layer that developers write to — then whoever owns or most deeply shapes that standard controls the integration tax of an industry whose applications we can’t fully specify yet.

The counterargument is that open standards, once donated to foundations and broadly adopted, don’t generate the same lock-in as proprietary platforms. OS/360 was IBM’s. CUDA is NVIDIA’s. MCP is now the Linux Foundation’s, with OpenAI and Google as co-stewards. The historical pattern suggests the moat accrues to whoever owns the layer, not whoever invented it.

Which may mean the next great platform play is still being assembled in a room we haven’t seen yet — the way IBM’s System/360 was being architected in a Connecticut motor lodge in 1961, three years before anyone else knew what was coming.

Categories
AI History

The Arrival

Yoshua Bengio spent forty years building the foundation of modern artificial intelligence. He won the Turing Award for it. And he didn’t think he’d live to see it work.

That’s the quiet fact buried inside Stephen Witt’s New Yorker profile of him. Bengio — one of the three researchers whose decades-long bet on neural networks eventually became the architecture underlying every large language model running today — had made peace with the idea that the thing he was building was a multi-generational project. Something for his successors to finish. Then Witt writes: “one day in late 2022, the technology had simply arrived. He compared it to meeting an extraterrestrial.”

Hemingway once described bankruptcy happening two ways: gradually, then suddenly. He meant ruin. Bengio experienced something harder to name — not ruin but arrival, which carries its own vertigo. The gradually was four decades of work that most of his peers considered quixotic. The suddenly was a Tuesday in November when a chat interface went live and the world quietly changed.

What unsettles me about the extraterrestrial comparison isn’t the strangeness it implies. It’s the distance. You meet an alien; you don’t meet something you made. The metaphor suggests that even its creator couldn’t fully recognize it — that the thing, once arrived, belonged to a category that exceeded its own origins.

We don’t have good language for this. Breakthrough, inflection point, paradigm shift — these are words people reach for after the fact, when they’re building timelines. What Bengio seems to be describing is the experience of standing in front of a threshold you spent your life approaching, and finding it already behind you.

The technology didn’t ask permission. It didn’t announce itself.

It arrived.

Categories
Business History Memories Radio

Permissionless Airwaves: The Legacy of FCC Part 15

Right now, as you read this, the air around you is thick with invisible conversations. Your phone is whispering to your router, your wireless headphones are singing to your laptop, and the smartwatch on your wrist is syncing quietly in the background.

We take this symphonic digital ecosystem completely for granted. But this panoply of wireless magic wasn’t just an inevitable product of technological march. It exists because of a profound, remarkably philosophical decision made by a bureaucracy in 1985.

It traces back to a seemingly mundane piece of regulatory code: the Federal Communications Commission’s Part 15 rules.

Historically, the airwaves were treated like highly exclusive real estate. If you wanted to broadcast a signal, you needed a license, a specific frequency, and a strict, government-approved mandate for what you were doing.

But within the radio spectrum, there were segments known as the ISM bands (Industrial, Scientific, and Medical). These were essentially the “garbage bands” of the airwaves. Microwave ovens, for instance, operated here, blasting out radio noise at 2.4 GHz. The interference was so heavy that the spectrum was considered practically useless for traditional communications.

Enter an FCC engineer named Michael Marcus. Marcus possessed a visionary understanding of a World War II-era technology called “spread spectrum” (famously co-invented by actress Hedy Lamarr). Spread spectrum didn’t rely on a single, clean channel; instead, it scattered a signal across a wide swath of frequencies, easily dodging interference.

Marcus argued for something radical: what if we opened up these “junk” bands to the public, allowing anyone to use spread-spectrum devices without asking for a license, so long as they adhered to basic power limits and didn’t cause harmful interference to primary users?

Incumbents fought it bitterly. Broadcasters and traditional telecommunications companies warned of absolute chaos. But in 1985, the FCC adopted the new Part 15 rules.

“We often talk about the great technological breakthroughs of our time as hardware or software triumphs. But sometimes, the most important enabling technology is just a clearing in the woods.”

Think about the nature of most regulation. It usually prescribes behavior. It looks at the future and says, “You may do exactly X, under condition Y.” But the Part 15 ruling did the opposite. It created a sandbox. The FCC didn’t try to predict Wi-Fi, Bluetooth, cordless phones, baby monitors, or the Internet of Things. In fact, they couldn’t have. They simply set the structural ground rules for how devices should coexist without stepping on each other’s toes, and then they stepped back.

This is the beauty of permissionless innovation. When you don’t have to ask a gatekeeper for access, a massive, uncoordinated burst of creativity happens.

A small company in the Netherlands could start working on what would eventually become Wi-Fi. Ericsson could invent Bluetooth. Innovators didn’t need to petition the government to launch a new product; the space was already cleared for them to play.

Part 15 was an admission of humility by a regulatory body—an acknowledgment that the most profound inventions are the ones we cannot yet foresee.

The greatest legacy of Part 15 isn’t Wi-Fi or Bluetooth. It is the enduring lesson that when you give brilliant minds a blank canvas and the freedom to experiment without asking permission, they will build a world more connected than you ever dared to imagine.

Note: this post was triggered by my reading of David Pogue’s new book Apple: The First 50 Years in which he describes the development of the Apple III and how its design met the requirements of the FCC’s Part 15 in terms of reduced RF interference.

Categories
Economics Energy History

Hoarding and Recession

While listening this afternoon to the latest edition of the Goodfellows podcast from the Hoover Institution I was reminded that I had been wanting to better understand the effects of hoarding behavior aggravating a situation such as the current conflict in the Middle East.

Wanting to explore this further I asked Grok the following: 

I’m listening to this discussion about recessions: https://youtu.be/UIB-fCsXzgs?si=L5kXmubqZOtcKnsz

Something I’ve wondered about in the present moment with the Iran situation is the effect of hoarding. 

What insights can you provide?

Here’s the response which I found very interesting:

Hoarding in the Iran Crisis: A Classic Shock Amplifier That Echoes the Podcast’s “Locusts and Pirates” Warning

You’re listening to exactly the right conversation. The Hoover Institution episode (just dropped today, March 25, 2026) with Tyler Goodspeed, Neil Ferguson, John Cochrane, and H.R. McMaster frames recessions as unpredictable “murders” by external shocks—locust plagues, pirate raids on shipping, coal strikes, or oil embargoes—rather than economies “dying of old age.”

The panel explicitly flags the ongoing Iran war as a live example: a potential energy shock via the Strait of Hormuz that could replay 1970s-style pain, especially if prolonged.

Hoarding is the behavioral multiplier they reference from history (panic coal buying in past shortages, 2020 toilet-paper runs), and right now it’s playing out in real time with the Iran situation.

Here’s a clear-eyed breakdown of its economic mechanics and recession implications.

The Current Iran Shock (as of March 25, 2026)

Since late February, U.S.-Israeli strikes on Iran (including the assassination of Supreme Leader Khamenei and hits on oil infrastructure) triggered Iranian retaliation: closure of the Strait of Hormuz, attacks on tankers, and disruption of ~20% of global seaborne oil and LNG. Oil prices spiked above $100/barrel (Brent briefly hit $120 in early March before partial pullback), European gas doubled in spots, and shipping has largely halted. The war is not over—Trump has signaled possible diplomacy, but Iran denies talks and continues missile activity—leaving markets in a state of fragile “wait-and-see” with elevated volatility.

This is precisely the kind of supply-side “Captain Hook” shock Goodspeed describes: physical disruption (not just fear) that raises input costs economy-wide.

How Hoarding Enters the Picture—and Why It Matters More Than People Realize

Hoarding here operates on three levels: national governments, corporations/refiners, and households. It’s rational self-protection in uncertainty, but collectively it turns a manageable disruption into a self-reinforcing spiral.

  • National-level hoarding (the big one right now): Asian importers (India, Bangladesh, Thailand, Philippines, Nepal, Pakistan) are already redirecting fuel from industry to households, imposing work-from-home orders, rationing cooking gas/LPG, shortening workweeks, and stationing troops at depots to prevent private hoarding. China—smartly—had pre-stocked ~1.4 billion barrels of crude before the war and is still buying aggressively. The result? A “mother of all bidding wars” for remaining supply, as one analyst put it. Even the U.S. and allies face pressure not to release Strategic Petroleum Reserves (SPR) too quickly, because everyone else is scrambling. IEA has ordered record releases, but if countries hoard reserves instead of sharing, global tightness worsens.
  • Corporate and trader hoarding: Refiners and traders are filling tanks wherever possible. Storage in the Gulf is hitting capacity, forcing producers to cut output (a real supply loss, not just transport). This adds non-fundamental demand and delays the normal market rebalancing.
  • Household panic buying: So far limited in the U.S./Europe (no widespread gas-line scenes yet), but it shows up as higher gasoline demand in Asia and could spread if prices keep climbing. The podcast’s historical note on 20th-century coal hoarding (“a woman begging for extra deliveries”) is spot-on for how this cascades.

The Recession Feedback Loop: Hoarding as an Accelerator

Here’s where it ties directly to the episode’s thesis:

  1. Price spike + inflation pass-through → Oil at $100+ for weeks/months fans headline inflation (0.4–1% globally per IMF models if sustained). Central banks face a dilemma: hike rates (worsening slowdown) or hold (letting inflation embed). The podcast notes 1970s price controls and windfall taxes amplified this exact dynamic.
  2. Demand destruction via higher costs → Households and businesses cut discretionary spending (travel, manufacturing, retail). Asian emerging markets—already rationing—are seeing this hit hardest and fastest. In the U.S., Goldman Sachs just raised recession odds to 25% citing the oil upside risk on top of softer jobs/retail data.
  3. Financial and confidence channel → Bidding wars drain liquidity, raise borrowing costs, and erode business investment. The episode highlights how supply shocks often trigger internal monetary drains (banks pull back lending) and equity drops—exactly what we’re seeing in modest S&P pullbacks and VIX spikes.
  4. Duration matters (the “prolonged” risk) → Short war = modest 10–30¢/gallon gas bump, limited recession odds. Prolonged Hormuz closure = guaranteed global recession via hoarding-driven tightness, per multiple analysts. Storage refill demand (U.S. SPR alone needs ~450 million barrels) adds extra buying pressure even after fighting stops.

Goodspeed’s optimism—that modern economies are more diversified and expansions last longer—still holds, but only if policymakers avoid 1970s mistakes (controls, hoarding of reserves). Strategic releases and diplomacy can short-circuit the hoarding spiral.

Bottom line: Hoarding isn’t the root cause (the Strait closure is), but it’s the behavioral “locust swarm” that turns a geopolitical event into a broader economic shock.

The podcast is prescient here—watch how SPR policy and Asian rationing play out in the next 2–4 weeks. If the war de-escalates quickly, hoarding fades and the recession scare passes.

If not, we’re living the exact multi-causal downturn the panel is warning about.

Categories
History Living Telephones

The Coiled Tether

Do you remember the physical weight of a conversation? It lived in the coiled, plastic spring of a landline telephone cord. We would stretch it across the kitchen, pacing over linoleum floors, the coil twisting around our fingers as we talked into the evening.

That cord was a literal tether. It confined us to a specific radius, but in doing so, it anchored us to the present moment. When you were on the phone, you were nowhere else. You were anchored to the wall, and by extension, to the person on the other end of the line.

There was also the sheer tactile satisfaction of the device itself—the heavy, contoured plastic of the receiver that fit perfectly between shoulder and ear, and the definitive, emphatic slam of hanging up on someone, a punctuation mark that the gentle tap of a touchscreen will never quite replicate.

Then came the subtle, sharp click on the line. Call waiting.

“We traded deep, uninterrupted connection for the anxiety of possibility.”

It was our first taste of modern conversational fragmentation.

Before call waiting, a busy signal was a polite “do not disturb” sign hung on the door of an ongoing dialogue. It meant you were occupied, engaged, entirely spoken for.

The click changed everything. It introduced a sudden, silent geometry to our relationships. When that secondary tone sounded, you were forced into a split-second hierarchy: do I stay with the person I am talking to, or do I chase the mystery of the unknown caller? The phrase, “Can you hold for a second?” became a small, culturally accepted betrayal of the present moment.

We traded deep, uninterrupted connection for the anxiety of possibility.

Eventually, the mystery of the ringing phone was solved altogether by a small, rectangular box with a glowing LCD screen: Caller ID.

For decades, a ringing phone was an invitation to a blind date. You picked up the receiver with a mix of anticipation and vulnerability. It could be a best friend, a wrong number, a telemarketer, or the person you’d been hoping would call all week. You answered with a universal greeting—a neutral, expectant “Hello?”—because you had no idea who was stepping into your home through the wire.

Caller ID gave us the power of the gatekeeper. It allowed us to screen, to prepare, to decide if we had the emotional bandwidth for the name flashing in digital text. We gained control, but we lost serendipity. We lost the unfiltered, genuine surprise of hearing a familiar voice when we least expected it. We stopped opening the door blindly and started looking through the peephole.

Today, we are entirely untethered. There are no coiled cords tying us to the kitchen wall. We carry our communication in our pockets, capable of ignoring texts, sending calls to voicemail, and managing our availability with unprecedented precision. Yet, for all this freedom and control, it often feels as though we are more disconnected than ever.

The good old days weren’t necessarily better because the technology was superior; they were beautiful because the limitations of the technology forced us to be human. The cord forced us to stay put. The lack of caller ID forced us to be open. The absence of call waiting forced us to finish the conversation we started.

Sometimes, looking back, I miss the simple, undeniable commitment of answering a ringing phone, twisting the cord around my index finger, and just listening.

Categories
Aging Citizens Band Radio History Living

The Static We Left Behind

There was a time when the airwaves crackled with a distinct, unpolished kind of magic. It wasn’t the curated broadcast of a corporate radio station, but the raw, spontaneous voices of strangers sharing the same lonely stretch of highway or suburban night. When I previously wrote about the rise and decline of CB radio, I didn’t fully anticipate how deeply the piece would resonate. The influx of emails, comments, and shared memories pointed to a singular, striking truth: we don’t just miss the hardware of the 1970s; we miss the serendipity of the connection it offered.

In the decades since the fiberglass whip antenna faded from the American automotive silhouette, our society has become infinitely more “connected.” We carry glass slabs in our pockets capable of reaching anyone, anywhere, in an instant. Yet, paradoxically, we often find ourselves feeling more profoundly isolated. The modern digital landscape is largely an algorithmic echo chamber, meticulously designed to feed us reflections of what we already know and who we already are.

CB radio, by contrast, was a geographic lottery. You turned the dial, adjusted the squelch, and were instantly thrust into a transient community composed entirely of whoever happened to be within your physical radius. It was messy, chaotic, occasionally absurd, and deeply human. It was a localized town square operating on a 27 MHz frequency.

“We traded the spontaneous for the scheduled. We swapped the local for the global… We traded the crackle of static for the endless, frictionless scroll of the feed.”

Reflecting on the quiet that eventually fell over Channel 19, it becomes clear that the decline of CB radio was more than just a technological shift—it was a cultural one. We traded the spontaneous for the scheduled. We swapped the local for the global, and the intimately anonymous for the hyper-public. We traded the crackle of static for the endless, frictionless scroll of the feed.

But the fundamental human impulse that fueled the CB craze never actually disappeared. The desire to reach out into the dark void and hear a human voice echo back—the spirit of “Breaker 1-9, is anyone out there?”—remains hardwired into our psychology. We see fragmented echoes of it today in late-night Reddit threads, in niche Discord servers, and in the fleeting, unscripted interactions of multiplayer gaming. We are all still, in our own ways, searching for a shared frequency.

Perhaps the true legacy of the CB radio isn’t a cautionary tale of obsolescence, but a gentle reminder. It reminds us that in our highly polished, curated digital world, there is still immense, undeniable value in the unscripted encounter. We haven’t lost the need to connect; we are simply navigating a world with too much noise and too few open channels.

Categories
AI History Work

Flash-Frozen Cognition: Birdseye, AI, and the Future of Work

I was listening recently to a conversation between Liz Thomas, Tom Lee, and Michael Lewis — the kind of wide-ranging dialogue where a single offhand story can suddenly anchor everything that’s been swirling loosely in your mind.

Tom’s story was about the 1930s, the weight of the Great Depression, and a man named Clarence Birdseye.

Birdseye had watched the Inuit fish in the brutal cold of Labrador and noticed something the rest of the world had missed: fish frozen instantly at sub-zero temperatures tasted perfectly fresh when thawed. The ice crystals formed too quickly to rupture the cellular walls of the flesh. He took that observation home, patented the process, and introduced the world to flash freezing.

On the surface, he had simply figured out a better way to keep peas green and fish edible. What he had actually done was detonate a quiet economic bomb.

Before Birdseye, entire ecosystems of seasonal labor existed to preserve, salt, can, and rush perishable goods to market before they rotted. When flash freezing arrived, those jobs didn’t evolve — they vanished. The ice harvesters, the seasonal canners, the local preservationists all felt the sudden, biting frost of obsolescence. The cold came fast, and it was indifferent.

Yet zoom out on the timeline, and a different picture emerges entirely. Flash freezing didn’t just kill jobs — it invented new ones that nobody could have anticipated. It necessitated refrigerated trucking. It transformed the grocery store, conjuring the frozen food aisle from nothing. It reshaped the home appliance industry, making the household freezer a fixture of modern life. Most profoundly, it decoupled humanity from the harsh dictates of the harvest season, democratizing access to nutrition across geographies and income levels that had never known that kind of abundance.

The destruction was visible and immediate. The creation was invisible and slow — and vastly larger.

Listening to Tom tell this story, I couldn’t help but see our own reflection in it.

Right now, we are all hyper-focused on the ice harvesters of the cognitive economy. We look at AI — large language models, generative tools, automated reasoning — and we see the rupture. We mourn the entry-level analyst, the copywriter, the junior coder. The anxiety is real. The displacement is real. The cold is real.

But what we are struggling to visualize is the refrigerated trucking of the mind.

“AI is flash-freezing cognition. It is taking tasks that used to rot if not attended to immediately by expensive, time-consuming human effort, and preserving them in a scalable, frictionless state.”

When intelligence and execution can be flash-frozen and shipped anywhere instantly — to a first-generation entrepreneur in rural India, to a solo founder with no budget for consultants, to a teacher in a school that can’t afford specialists — what new aisles get built in the supermarket of human endeavor?

The honest answer is that we don’t know. The Inuit fishermen of Labrador couldn’t have imagined the frozen pizza aisle. The ice harvesters of the 1930s couldn’t have pictured the cold chain logistics industry that employs millions today. We are standing in their moment, watching the ice form, mourning the harvest — and almost certainly underestimating what comes next.

The true impact of AI won’t be measured in the jobs it automates. It will be measured in the industries, creative liberties, and human possibilities that emerge because we no longer have to spend all our energy just keeping the ideas from spoiling.

Questions to Consider

  1. The Invisible Creation: Flash freezing’s job creation vastly outpaced its job destruction — but only over decades. How long are we willing to hold that faith with AI, and what do we owe the people displaced in the interim?
  2. The Democratization Dividend: Birdseye’s invention ultimately made fresh nutrition available to people who never had it. Who are the equivalent beneficiaries of flash-frozen cognition — and are we building the infrastructure to actually reach them?
  3. The Harvest Season Question: We’ve always structured education, careers, and institutions around the assumption that expertise is scarce and slow to develop. What breaks — and what gets liberated — when that assumption stops being true?
  4. The Indifference Problem: The cold that killed the ice harvesters’ livelihoods was indifferent to their suffering. Is there anything about AI disruption that is meaningfully different from previous waves of technological displacement — or are we simply the latest generation to stand in that frost?

Categories
AI Farming History

The Harvest and the Algorithm: What 1990s Farms Teach Us About AI

There’s a strange kind of wisdom hiding in dusty old books about agriculture.

When you’re caught in the middle of a technological revolution—and with AI, there’s no question that we are—it’s tempting to keep your eyes fixed on the horizon. But sometimes the most clarifying thing you can do is look back.

Tracy Alloway at Bloomberg recently pointed to something genuinely instructive from the past: Richard Critchfield’s 1990 book, Trees, Why Do You Wait? America’s Changing Rural Culture, which traced the collapse of the family farm as industrial agriculture swept through the Midwest.

The broad strokes are familiar. As machinery got more expensive and efficiency became everything, scale won. The 80-acre husband-and-wife operation got swallowed by the 2,000-acre neighbor with access to capital. It wasn’t complicated. It was just gravity.

But here’s the part that should make your ears prick up.

The Seed That Was Supposed to Save Everyone

In the late 1980s, agricultural biotechnology arrived with a very specific promise. The idea was almost elegant: if you could bake the magic directly into the seed, you wouldn’t need all that expensive machinery, all those sprawling acres, all that fertilizer. The playing field would tilt back toward the small farmer.

Critchfield quoted an Office of Technology Assessment report from 1986 that captured the mood of the moment:

“The Office of Technology Assessment in 1986 forecast that biotechnology in crops would be more quickly adopted by richer farmers… Others argue that the more that gets built into the seed itself, the more it means higher yields at lower cost… If it reduced farm income, it could work to the smaller farmer’s advantage. As it is with all new technology, it is hard to foresee the consequences.”

You can feel the cautious optimism in that language. Hard to foresee the consequences. The understatement of a century.

What Actually Happened

The biotech did raise yields. Nobody disputes that. What it didn’t do was leave the gains in the hands of the people doing the actual farming.

Thanks to intellectual property law, patent protections, and a level of corporate consolidation that would have seemed cartoonish if you’d predicted it in advance, the value flowed straight upstream. We didn’t get “more in the seed, less paid for inputs.” We got more in the seed, and vastly more paid for proprietary inputs. The tech giants of agriculture captured the surplus. The farmers got the risk.

Now Listen to How We Talk About AI

We are told AI will democratize expertise. That a one-person startup will be able to code like a ten-person engineering team. That a small business will generate world-class marketing copy. That this is, finally, the great leveler.

Sound familiar?

Alloway’s analysis lands hard precisely because it forces the uncomfortable question: who will actually capture this value? The ownership structure of AI looks eerily similar to the agricultural biotech boom—proprietary models, walled-off training data, and a handful of enormous tech companies positioned to act as tollbooths between everyone else and their own productivity gains.

She’s right to note that “the ultimate distribution of benefits isn’t determined by technology alone. Policy also plays a role.” That sentence is doing a lot of quiet work.

If the agricultural analogy holds, productivity gains from AI won’t naturally flow to the individual worker or the small business owner. Without a robust open-source ecosystem or some deliberate policy intervention, those gains will be captured by whoever controls the compute and the models.

Where the Analogy Might Break Down

Here’s where I think there’s room for genuine optimism—not naive optimism, but structurally grounded optimism.

You cannot open-source arable land. Reverse-engineering a patented biological seed is genuinely hard, legally risky, and practically difficult. Code and model weights are different. They’re infinitely replicable. The marginal cost of distribution is essentially zero.

The battle between closed, proprietary AI and open-source models is still very much live. That’s not nothing. AI is fundamentally more commoditizable than a physical farm, and the history of software suggests that open ecosystems have a real shot when the community is motivated enough to build them.

Who Owns the Harvest?

Technology can reshape daily workflows in months. Power structures take decades to budge, if they budge at all. The mistake would be assuming the former automatically changes the latter.

The question worth sitting with isn’t what can AI do—that list gets longer every week. The question is who decides how the productivity it unlocks gets distributed. That’s not an algorithm problem. It’s a political and economic one.

If we want the AI revolution to be a rising tide rather than another tractor paving over the family farm, we have to look past the technology itself. We have to decide, deliberately, who owns the harvest.

Questions to Ponder

On history and pattern recognition: The agricultural biotech optimists weren’t stupid—they were looking at the technology and making reasonable inferences. What does that tell us about the limits of predicting who benefits from a new technology by studying the technology itself?

On open source as a counterweight: The open-source AI movement (Llama, Mistral, DeepSeek) is often framed as a technical story. Should we be thinking about it primarily as a political economy story—a structural check on proprietary capture?

On the role of policy: Antitrust law, data ownership rights, compute access regulation—which levers, if any, seem realistic? And who has the incentive to pull them?

On the worker vs. the firm: If AI raises individual productivity, does the gain show up in wages, prices, profits, or somewhere else? What would need to be true for workers to actually keep a meaningful share?

On commoditization speed: Software and model weights can be replicated freely—but does speed matter? If proprietary models establish deep lock-in before open alternatives mature, does the theoretical commoditizability even help?

Inspired by Tracy Alloway’s analysis at Bloomberg and Richard Critchfield’s Trees, Why Do You Wait? (1990)

Categories
Authors Books History

The Devil’s Rope

We often mistake simplicity for innocence. When we look at a technological innovation, we tend to judge its weight by its complexity—the microchip, the steam engine, the nuclear reactor. But sometimes, history turns on the axis of something far more rudimentary. Sometimes, the world changes not with a bang, but with a sharp, metallic scratch.

I was recently reading Cattle Kingdom by Christopher Knowlton, and I stopped cold at a passage regarding the invention of barbed wire. It’s an object we pass by on highways or stumble over in overgrown fields without a second thought. Yet, Knowlton writes:

“None was more significant than the creation of barbed wire, which literally reshaped the landscape and set the stage for the era’s eventual destruction—at great personal cost to so many of its key players.”

It is a profound observation. We tend to romanticize the American West as a geography of endless horizons—a place defined by what it didn’t have: fences, borders, limits. It was the Open Range. But that openness was fragile. It existed only as long as the technology to close it was absent.

When Joseph Glidden and others patented their variations of “The Devil’s Rope” in the 1870s, they weren’t just selling steel fencing; they were selling a new concept of ownership. Before wire, a man owned what he could patrol. After wire, a man owned what he could enclose.

The quote strikes a melancholic chord because it highlights a paradox of human progress: the tool created to maximize the land ended up destroying the culture that relied on it. The cowboys, the cattle barons, and the drifters who defined the era were undone by the very efficiency they sought. The wire made the cattle industry profitable on a massive scale, but it also ended the cowboy’s way of life. It stopped the long drives. It turned the cowboy from a navigator of the plains into a gatekeeper.

And, as Knowlton notes, the “personal cost” was staggering. This reshaping of the landscape wasn’t just aesthetic; it was violent. The wire cut off migration routes for bison and the Indigenous tribes who followed them. It sparked the fence-cutting wars, neighbor turning against neighbor in the dark of night, snapping tension wires that represented their livelihood or their imprisonment, depending on which side of the post they stood.

There is a lesson here for us today, far removed from the dusty plains. We are constantly inventing our own versions of barbed wire—digital boundaries, algorithmic silos, tools designed to corral information or efficiency. We build these structures to create order, to claim our stake, and to protect what is ours. But every time we draw a line, we must ask: what era are we destroying? What open range are we closing off forever?

The landscape is always being reshaped. The question is whether we are building fences that protect us, or cages that trap us in.

Categories
AI History Living

The Echo of the Roar

It is a strange sensation to look back exactly one century and see our own reflection staring back at us, sepia-toned but unmistakably familiar. We often think of the “Roaring Twenties” as a stylistic era—flapper dresses, Art Deco skyscrapers, and jazz. But beneath the aesthetic was a seismic technological shift that mirrors our current moment with an almost eerie precision.

In the 1920s, the world was shrinking. The radio was the “Great Disrupter” of the day. For the first time in human history, a voice could travel instantly from a studio in Pittsburgh to a farm in Nebraska. It was the democratization of information, a sudden collapse of distance that left society both thrilled and anxious.

“The radio brought the world into the living room; the algorithm brings the universe into our pockets.”

Today, we stand in the wash of a similar wave. If the radio brought the world into the living room, the internet—and specifically the generative AI of this decade—has brought the collective consciousness of humanity into our pockets.

The parallels in infrastructure are just as striking. One hundred years ago, the internal combustion engine was reshaping the physical landscape. The horse was yielding to the Model T; mud paths were being paved into highways. The very geography of how we lived was being rewritten by the automobile. In the 2020s, the “highway” is digital, built on cloud infrastructure and fiber optics, and the vehicle isn’t a Ford, but an algorithm. We are transitioning from physical labor to cognitive automation just as they transitioned from animal labor to mechanical muscle.

The Texture of Time

There is a specific texture to this kind of time. It is a mix of vertigo and acceleration. In 1925, the cultural critic might have worried that the “machine age” was stripping away our humanity, turning men into cogs on an assembly line. In 2025, we worry that the “algorithmic age” is stripping away our agency, turning creativity into a prompt.

But here is the insight that offers me comfort: The 1920s were chaotic, yes, but they were also a crucible of immense creativity. The pressure of that technological change forged modernism in literature, new forms of architecture, and entirely new ways of understanding the universe (quantum mechanics began finding its footing then).

We are not just passive observers of a repeating cycle. We are the navigators of the rhyme. The technology changes—from vacuum tubes to neural networks—but the human task remains the same: to find the signal in the static. To ensure that as the machines get faster, our souls do not merely get cheaper. We must decide, just as they had to a century ago, whether we will be consumed by the roar, or if we will learn to conduct the music.