Categories
San Francisco/California Writing

The Company of Fog

Note: a little fun this morning with a piece envisioning Jimmy Breslin writing about lunch with Herb Caen at Tadich Grill…

I went to San Francisco to meet a man who writes about the place the way a guy writes about a woman he still wants to sleep with after thirty years.

His name is Herb Caen. He has been telling this city about itself every morning since before most of the people reading him were born. They call him Mr. San Francisco. In New York we don’t hand out titles like that. In New York you earn the right to be called a son of a bitch, and you take it as a compliment.

We met at Tadich Grill. Open since the gold rush, and it still looks like it never got the news that things change. Dark wood. White tablecloths. Waiters old enough to remember when the bridge was new and still sore about it. The fish comes out of the water and onto the plate without a lot of conversation in between.

Caen was already sitting when I got there. Neat. Jacket. That California tan that never quite goes away, like the sun signed a contract with him. He stood up and shook my hand like a man who has shaken a lot of hands and still means it. I ordered a martini, because when you’re in another man’s town you drink what the town drinks, even when the town is wrong.

He had one too. Stoli, twist. He called it Vitamin V.

We talked about columns. About writing for a city that never sleeps against writing for one that takes long naps in the fog and calls it lifestyle. I told him New York will rob you and hold the door on the way out. He said San Francisco smiles while it’s doing it, then writes a nice item about your necktie.

He works in three dots. Little pieces, one after another, like a man feeding a parking meter all day instead of buying the car. Gossip, jokes, the name of whoever just opened a place on Union Street. I need more room than that. I need to turn around. He said my way was fine if you like carrying a piano up the stairs every day. I told him his was fine if you don’t mind never sitting down long enough to find out what you actually think.

The sand dabs came. Caen looked at his like a priest looks at the host. I ate mine like a man who skipped breakfast. Everything on that table was good — the bread, the butter, the light coming in off the street looking like it had been there as long as the wood and knew it.

He leaned in. “You know the difference between our two cities?”

I waited.

“In New York, the people are the story. Here, the city is the story. The people just get to live in it for a while.”

That’s a hell of a thing for a man to say about his own hometown — that the place is bigger than everybody in it, including him. I told him about the man who dug Kennedy’s grave, the one nobody else thought to ask. Caen nodded. He understood the assignment. It was the only time all lunch he didn’t have a joke ready.

We stayed too long. The martinis did their work. Outside, the fog came in soft and certain, the way it does out there, like it’s been rehearsing. Caen put his coat on like a man who has never once been in a hurry. I put mine on like a man who has somewhere else to be, even when he doesn’t.

On the sidewalk he tipped his hat. An actual hat. Nobody tips a hat in New York anymore. We’re too busy hailing something.

“Come back,” he said. “The city likes company.”

I walked toward the cable car line, the fog thick enough to make the streetlights float like they’d given up on the ground. Two men, two ways of loving a place. One writes it in little pieces that catch light for a second and go dark before the next one starts. The other digs until he finds the one man nobody else remembered to ask about, and doesn’t stop until the shovel hits something.

Neither of us was going to change. Neither was the city. That’s probably why we got along.

I caught the cable car and rode it a few blocks just to feel it move under me, the way you touch a thing to make sure it’s real. Behind me somewhere, a man who’d spent his life telling San Francisco it was beautiful was walking home through the fog he’d helped invent — which is either the best job in America or a life sentence, and he never once let on which.

I never did find out if he got tired of saying the same true thing a different way every morning. I didn’t ask. Some men you let keep their secret. Besides, I had my own city waiting, and it doesn’t smile at you while it robs you. It just takes the money and tells you to have a nice day.

Categories
AI AI: Prompting

The Price of the Barstool

Some genius out in California — one of the AI guys, the smart ones, always the smart ones — says the trick to getting a machine to understand you is to stop typing and start talking. Ramble for ten minutes, he says. Total mess. Say whatever’s in your head, contradict yourself, circle back, don’t clean it up. The machine, he says, is better at finding what you meant than you are.

I could’ve told him that thirty years ago for the price of a cup of coffee, except I would’ve told him to go find a bartender.

Every good bartender in this city has been doing this since before anybody could spell computer. You sit down, you’re a mess, you talk for twenty minutes about your ex-wife and your kid’s tuition and the guy at work who’s getting the promotion you should’ve gotten, and it comes out in no particular order, and somewhere in the fourth minute you say the one true thing — I think I’m afraid I already peaked — and the bartender doesn’t say a word, just wipes the bar down in front of you, and by the time you leave you know something about yourself you didn’t know when you walked in. Nobody paid Sam a nickel for that. Nobody gave him a paper to publish.

Now they’ve built a machine that does the same trick and they’re calling it a pattern. They put it in a paper. Somebody in Palo Alto’s going to raise money on it.

Here’s what I’ll say for the guy — he’s right, and being right is rarer than people think in that business. The mess is where the truth lives. Nobody ever said something true the first time they tried to say it cleanly. You clean it up too fast, you’ve written a memo. You let it run messy, you’ve said something.

But I’ll tell you what he won’t put in the paper, because it doesn’t fit on a slide. The machine will hand you back a cleaner version of your ten minutes, and it’ll sound better than what you said, and half the time it’ll be missing the one part that was actually true, because the true part usually comes out sounding wrong the first time. That’s the part a bartender knows and a machine doesn’t. Sam wouldn’t have cleaned up your sentence. He’d have just remembered you said it, and brought you another one, and let you sit there with it.

They’ll charge you a subscription for the part where it listens. The part where somebody sits with what you actually said — that’s still free, if you can find the right stool.

Categories
Aviation Memories

Overhead the Farm

This week is the annual EAA Fly-In in Oshkosh, Wisconsin. This brings back memories of a good time we shared over 35 years ago.

Our Piper Cherokee could hold enough fuel to fly well past what our bladders would tolerate, but we didn’t fly that way. We planned every stop — fuel first, restrooms too — and let the country reveal itself in the increments a small plane gives you: Rock Springs WY for the first night, Alliance NE next for gas and bathroom, Yankton SD to make a phone call, and then, finally, circling over a farm outside Alcester, South Dakota, that I hadn’t seen from the air before.

It was 1989. I was forty-two. My Dad was sixty-eight, sitting in the right seat of my 1974 Piper Pathfinder, painted red, white, and blue — an honest airplane that never asked for much and never gave us trouble. We were headed to Oshkosh, to the fly-in, the way pilots go to Mecca — except this time the pilgrimage ran straight through his childhood.

His people were Norwegian. His grandfather had crossed an ocean and then a good part of a continent to end up farming eastern South Dakota, and the family had stayed put the way families did then, generation folding into generation until my grandfather had more children than the land could easily absorb. Some of them farmed anyway. My father did not. He went into printing — Linotype machines, hot lead, the smell of a shop instead of the smell of a barn. But I remember the soil anyway. I’d visited that farm as a boy of ten or so, and what stayed with me wasn’t a story anyone told but a set of smells: hay and diesel and animal warmth inside the barn, the particular rattle of a tractor under me, dust rising gold in a shaft of light through a barn door left open. Did I want to help milk the cows?

We landed at Yankton for fuel and my father called his sister. I don’t remember what either of them said — it was a short call, the kind you make when you’re about to do something better shown than explained. Something like: we’ll be overhead in twenty minutes. Go outside. Look up.

And then we were. Alcester from a thousand feet is not much — a scatter of buildings, a windbreak of trees, fields squared off in the pattern only farmland makes — but down there was a farmhouse, and out in the yard were two people who’d stopped whatever they were doing to come out and watch the sky. We circled twice. My father waved — big, unmistakable — from a thousand feet up at his sister and her husband in the yard below. They waved back, two small shapes going back and forth, back and forth. Then we leveled the wings and kept going east, toward Wisconsin, and neither of us said much of anything, because there wasn’t really anything to say. It was just good.

Oshkosh itself is its own kind of pilgrimage, and if you’ve never been, the thing that’s hard to explain is not the airplanes — though there are more in one place than you will ever otherwise see — it’s the people.

And the crowd starts in the skies as you line up for the Fisk arrival into Oshkosh. The air traffic controllers are out in a field somewhere below and they’re talking continuously. “The red white and blue Cherokee rock your wings! Land on the yellow dot midfield…” Before you know it you’re turning off into the grass and being waved to your parking spot next to so many other airplanes.

Most pilots who fly in tent-camp right under their own wings, sleeping on the grass next to the machine that brought them. Dad and I opted for relative luxury: a dorm room at a local university, actual beds, actual plumbing. I don’t remember the room. Or how we got back and forth to the airport.

What I do remember is the grounds. Tens of thousands of people for a solid week, and no litter. None. If you saw a gum wrapper, you picked it up, the way you’d pick up a dropped tool in someone else’s shop. The shared bathrooms stayed clean not because anyone was assigned to clean them, but because everyone behaved as if they were company in someone else’s house.

It sounds like a small thing to remember thirty-some years later. It isn’t. I’ve described it as renewing my faith in humanity. Each afternoon we gathered along the flight line for the daily air show which was usually punctuated by the contrast between an old biplane and the sound of the latest American fighter aircraft. Amazing sights.

I had been once before, driven in with pilots who’d rented an RV, and learned mostly that I didn’t want to do that again — too many people, too close together, none of the ease of arriving on your own wings. That trip didn’t have the thing this one had. I couldn’t have told you what the thing was, exactly, except that being there with my Dad restored some baseline faith in how people can behave toward each other when they all care about the same unlikely thing.

The flight home took us back through Spearfish, where another of our relatives put us up for the night, and then on to Logan, Utah, for fuel and that bio break.

It was the leg after that — the crossing of the state of Nevada — that I still think about on rough days. The sky was clear. Not a cloud anywhere. And the air was some of the worst turbulence I’ve flown through, before or since. I tried up high, I tried down low, but there was no altitude in that whole stretch of sky that gave us any peace. No smooth air anywhere. Just wind. Dad didn’t get sick. Neither did I. We just held on and let the airplane buck and waited it out, the way you do.

After a time then we crossed low over the Sierras, and somewhere on the descent into the central valley of California the air went calm again, smooth as the farm pond back in Alcester must have been that morning, and the trip was, in every way that mattered, already over.

Categories
Writing

Wait, What? The Art of the Irresistible Lead

A few weeks ago I sat with a fresh draft of one of my daily reports, the cursor blinking on a perfectly serviceable first sentence. It told the reader exactly what the piece was about. It was clear. It was polite. And it was, frankly, a little boring.

I deleted it and tried again. Then again. On the fifth try, something shifted. The sentence wasn’t flashy, but it carried a quiet spark of surprise—an unexpected detail that made even me lean in. In that moment I felt the familiar thrill of a good lead doing its quiet work: not demanding attention, but inviting it.

Susan Orlean has long been one of my quiet heroes in this regard. She crafts first sentences that land like gentle provocations—declarative, sometimes absurd, always authoritative in their own small way. They dare you not to keep reading, even when the subject seems, at first glance, ordinary.

For years now I’ve paid close attention to these opening moves—not just in Orlean’s work, but across the writers who consistently pull me in and refuse to let go. In a world of endless distractions and shorter attention spans, the lead is no longer optional; it is an act of hospitality. It says, gently but firmly: Come sit with this for a while. There’s something worth your time here.

This post is my reflection on that craft—why it matters, what the masters do so well, and how those of us who write (whether for blogs, reports, letters, or larger ambitions) might get better at it. Along the way I’ll share specific examples that still make me pause and smile, plus some practical ways I’ve tried to strengthen my own first sentences. Because if curiosity and receptivity are part of my compass, then learning to write better leads feels like one small, ongoing way to live into that.

Why Leads Matter More Than Ever

Attention is the scarcest resource we have. A strong lead doesn’t just inform—it seduces, surprises, or jolts the reader into “Wait, what?” mode. It earns those first crucial moments in a distracted world.

When I struggle with a draft, it’s often because the opening is doing too much summarizing and not enough inviting. A great lead respects the reader’s time while promising payoff. It’s generous. And in my own writing, whether a morning report or a more personal essay, refining the lead has often clarified the entire piece.

Masters of the Lead

Over time I’ve noticed how the same principles appear across very different forms. Feature writers like Susan Orlean seduce us into caring about subjects we didn’t know existed. Obituary writers work under tighter constraints but with profound stakes—distilling a life into something that honors the dead while engaging the living.

Orlean often opens with vivid, affectionate character portraits or surprising details that feel like delightful invitations. One classic: “In Bulgaria, some tennis balls are like dumplings.” It’s playful, specific, and instantly makes you wonder about the world these athletes inhabit. Another strength is the rapid, layered sketch: “John Laroche is a tall guy, skinny as a stick, pale-eyed, slouch-shouldered, and sharply handsome, in spite of the fact that he is missing all his front teeth.” The list is cinematic; the twist humanizes. Orlean writes the way a reader experiences the story, starting at the beginning and pulling you along with warm observation.

Obituary writers like Alden Whitman show the form’s power under pressure. Whitman conducted thoughtful advance interviews, framing them as biographical updates rather than morbid exercises. He asked reflective questions about identity, turning points, and what mattered most—gathering the intimate details that made his leads gracefully revealing. His portraits treated lives as coherent stories, full of flavor and humanity. Modern masters like Margalit Fox continue this with witty, precise openings that capture essence in one surprising stroke. British obituary traditions add elegant wit and anecdote, often using a single eccentricity to humanize even formidable figures.

The contrast is illuminating. Orlean’s warmth draws you in with delight and empathy. Joan Didion might open the same subject with cool cultural dissection, noting contradictions and performance. Wright Thompson would make it mythic and sensory. Jimmy Breslin would give it street-smart heart and working-man respect.

I sometimes test this by imagining how each would open a scene I actually remember: Moe’s Books on Telegraph Avenue in 1969. I was young, new to Berkeley, and the place felt like a sanctuary. The cigar smoke drifted out onto the sidewalk. Moe himself held quiet authority behind the counter. Browsers moved slowly, as if time among books was time well stolen. He bought books fairly—no haggling, just a fair price—and held court with that cigar while the avenue hummed with protest outside.

Orlean might begin with affectionate observation: the smoke, the quiet authority, the browsers moving as if time were something you could borrow among the shelves. Didion would coolly note the cultural performance—the revolutionary texts amid fading energy. Thompson would immerse you in the sensory sanctuary, the myth of the place. Breslin would tell it like a guy at the bar: Moe buying books fairly, no nonsense, holding court with that cigar while the world outside churned.

Each approach turns the same details—cigars, fair dealing, the charged atmosphere—into something that hooks and reveals. What unites them is deep curiosity and receptivity: seeing people and places fully, with compassionate attention.

How to Get Good at Leads

Mastery comes from deliberate practice. Here are approaches I’ve found helpful:

  • Read and dissect. Keep a notebook of great leads. Note what emotion or curiosity they trigger and what promise they make.
  • Write multiples. For any story, draft five versions—zinger, scene, metaphor, question, quiet setup—then choose or combine.
  • Draft first, revise last. Write the piece, then craft the lead. Or iterate after you know the emotional core. Read aloud.
  • Hunt arresting details. Look for surprising combinations (Orlean) or telling contradictions (Didion). Your own memories are full of them.
  • Get feedback early. Share just the lead. Ask: Does this make you want to read more?
  • Experiment with constraints. Try ultra-short leads or imitate a master’s voice, then adapt to yours.

Over time this becomes instinctive. Leads force clarity about what your story really is.

Common Pitfalls

Avoid summaries that feel polite but boring, or cleverness that sacrifices clarity. The best leads align with your internal compass—generous, empathetic, curious. They reflect compassionate honesty: offering truth with care.

Writing better leads is one small way to live my life theme more fully. It’s about paying close attention, finding the spark in the everyday, and inviting others to do the same.

These values are my compass—and today, curiosity feels like the right direction.

Categories
Computers IBM

The Day the Last Mainframe Went Dark

Note: I literally grew up during the heyday of the IBM mainframe era. My first real job was working for IBM in San Francisco beginning in 1968. I’m a “big iron” kind of guy. But this post was imagined after reading the following in a July 14, 2026 announcement from IBM: When we discussed our expectations with you in April, we noted that we would be wrapping on the launch of z17 in the second quarter. Given this was the strongest start to a mainframe program in our history, we expected Infrastructure revenue to decline low-single digits for the year, beginning this quarter. What played out was worse than our expectations, driven by a shortfall in our Z performance and the associated software stack, primarily in Transaction Processing. In the last few weeks of June, we saw clients shift their quarterly capex spend toward servers, storage, and memory purchases to secure supply-constrained infrastructure ahead of expected price increases. This dynamic impacted client buying patterns. While we anticipated some supply chain related impact in our expectations, we did not anticipate the magnitude of the capex reprioritization.

It won’t arrive with fanfare. No countdown, no viral video of engineers raising a glass. One morning—sometime in the 2040s, perhaps later—a small team in a climate-controlled data center will complete the final cutover. They’ll flip the switches, watch the lights dim, and listen as the hum of the last production IBM mainframe fades to silence. An era that began with the System/360 in 1964 will end. Not with a crash. With the soft click of obsolescence.

We’ve been predicting the mainframe’s death for decades. In the early 1990s, pundits declared it doomed. They were wrong. Those systems—reliable, secure, capable of staggering transaction volumes with near-perfect uptime—became the invisible backbone of modern life. Your last bank transfer, airline reservation, insurance claim, or government benefit likely touched one. They endured because they solved hard problems well: high-volume, mission-critical processing where failure was never an option.

The path to that final power-down was never a rupture. It was a long, uneven evolution—driven by economics, technology, talent shifts, and the patient work of modernization. AI tools accelerated the transition. They didn’t cause it.

Lessons from Earlier Transitions

Steam engines dominated railroads for generations—powerful, reliable, deeply integrated into the industrial economy. Diesel won through incremental advantages: better efficiency, lower maintenance, longer trains with less labor. Railroads rebuilt infrastructure and retired the old iron as the economics aligned, route by route.

Prop planes opened the skies to mass travel. Jets brought speed and range that transformed global connectivity—but airlines didn’t scrap fleets overnight. They ran hybrids during the overlap, invested in new airports and training, and retired props as jet economics and passenger demand made the case irresistible.

The mainframe followed this pattern. AI coding agents—Claude Code, Cursor, OpenAI models, AWS Transform, IBM watsonx—transformed the brutal manual work of understanding undocumented COBOL, extracting buried business logic, generating tests, refactoring safely. What once demanded scarce veteran experts for months or years could now be accelerated, with rigorous human oversight and equivalence testing.

Platforms like Visa’s Pismo showed a smarter path: incremental modernization. Cloud-native microservices layered alongside legacy cores, rather than rip-and-replace. Banks demonstrated real progress. Hybrid strategies won—AI inference running close to sensitive data on evolved mainframes (IBM’s z17 and successors, with on-chip accelerators), while new applications and analytics moved to elastic cloud environments.

IBM positioned the platform as an “AI factory” for low-latency, secure workloads. But pricing pressure was constant. High, capacity-based software licensing made the economics harder to defend as cloud offered predictable, usage-driven costs and younger talent gravitated toward modern stacks. For CFOs weighing rising maintenance against retiring COBOL expertise and AI-assisted migration, the scales tipped.

By the mid-2030s, competitive and regulatory forces intensified. Fujitsu’s exit from mainframes created a cliff in affected markets. Skills shortages accelerated. Even the most conservative holdouts—ultra-high-volume, regulated systems in finance, government, specialized industries—began serious moves, as simulation environments and exhaustive parallel testing brought the risk down to manageable size.

The Final Act

The last systems to go were the stubborn ones, where disruption carried outsized consequences. When the final cutover succeeded—after months of flawless parallel running—the team powered down the machine. A global bank, a payments processor, a government entity. Maybe a small ceremony: engineers who’d kept it alive for decades, trading stories of the iron that never failed when the world needed it most.

Picture the aircraft boneyards outside Tucson or Victorville, retired 747s sitting in rows under the sun, giving up parts to new generations before they’re recycled. Mainframes will meet a similar fate, more climate-controlled. Some linger in warehouses as insurance, still humming faintly in test or archival roles. Others get dismantled by IT asset disposition teams—data wiped to standard, processors and I/O cards harvested for niche markets. The bulk gets recycled, metal and circuitry returning to the supply chain. Like the jets, the iron won’t vanish in disgrace. Its lessons in reliability and disciplined engineering at scale live on, embedded in whatever comes next.

The world didn’t stop. Transactions kept flowing, now on distributed, elastic, AI-augmented platforms that had absorbed the best of what came before. The mainframe era didn’t end in failure. It ended because better options finally existed for every workload.

What Endures

We’ll look back with respect and nostalgia. The mainframe wasn’t flashy, but it taught something durable: some problems reward obsessive focus on reliability and scale; disciplined engineering outlasts hype cycles; the wisest transitions are rarely clean breaks. They’re patient evolutions that carry forward what matters.

IBM will have completed its own transformation by then—software, services, hybrid orchestration, AI tools that work across environments. The company that built the platform helps close the book on it.

The last mainframe going dark won’t feel like loss. It will feel like the natural close of a chapter that powered the digital economy through its most formative decades. The iron did its job. Now the next architecture takes the stage, standing on shoulders built to last.

Categories
AI

The Kitchen, Not the Farm

There is a sentence buried in Thinking Machines Lab’s release notes for Inkling, its first proprietary model, that most companies would never let out the door. Describing their own creation, the company states plainly that Inkling is “not the strongest overall model available today, open or closed.”

Read that again. A startup that raised two billion dollars in seed funding at a twelve-billion-dollar valuation, founded by OpenAI’s former CTO and staffed with veterans of the labs currently locked in the most capital-intensive arms race in corporate history, shipped its debut model with an admission of inferiority attached to the label. Not buried in a footnote. Stated in the announcement.

It seems like this was the clearest signal yet that the frontier-capability race may be the wrong game, and that durable value in enterprise AI accrues not to whoever has the smartest model, but to whoever owns the layer where that model gets adapted to a particular customer’s purpose.

As I’ve thought about it, the AI industry seems to be stratifying into three distinct businesses, each with different economics, occupied by a different cast of companies.

It begins with the farm, where the raw ingredients get grown. Then there’s the kitchen, the capital equipment that makes skilled cooking possible at scale. Lastly there’s the restaurant, where somebody who understands a specific customer takes the ingredients, uses the kitchen, and puts a particular dish in front of a particular diner who is paying for a complete meal, not just the flour or the vegetables. Thinking Machines seems to me like a clear example of a company trying to explain which of those businesses it’s actually in. It is not the only one.

The sequence, read backward

Founded in February 2025. Silent for over a year. Then, last October, the company’s first product emerged — and it wasn’t a chatbot, wasn’t an assistant, wasn’t anything a consumer like me would recognize or understand. It was Tinker, a fine-tuning API. Infrastructure for customizing other people’s models, shipped before the company had released a model of its own.

That sequencing is the tell. A company chasing frontier supremacy builds the model first and the tooling around it later, the way the frontier AI labs have all done. Thinking Machines inverted the order. It built the workshop before it built anything to put in the workshop window, which only makes sense if the workshop was always the product.

Inkling, released this month, doesn’t reverse that logic. It completes it. The model is described in the company’s own materials as “an extremely knowledgeable, generalist base that can be extended via fine-tuning” — language that positions the model itself as raw material, not a finished good. It ships with full open weights, day-zero availability on Tinker, and a name chosen, according to the company, to evoke “an idea in its earliest stage, with the potential to grow into something greater.” Even the naming is a thesis statement. Inkling is not meant to be the only thing you use. It’s meant to be the thing you start from.

In the farm-kitchen-restaurant frame, it seems like Thinking Machines is trying to own two levels of the stack at once. Inkling is the farm — grown at real expense, forty-five trillion tokens of training data, frontier-scale compute. Tinker is the kitchen — the induction range and the walk-in fridge, sold as a service to whoever wants to cook. What Thinking Machines has explicitly declined to be, by its own admission, is the restaurant. They are not trying to serve you the best possible dish. They are trying to make sure that whoever does serve you that dish is buying their ingredients and standing at their stove and cooking in their kitchen.

The manifesto that preceded the model

A company doesn’t back into a strategy this coherent by accident. Earlier this month — before Inkling shipped — the lab published a position paper arguing that most AI today is trained in a handful of places and then frozen, a design that by its nature excludes the people the model is meant to serve. Their proposed alternative: AI that is distributed, customizable, and shaped by the people using it, not the lab that built it.

Mira Murati has said the same thing more plainly, and said it a year before Inkling existed, back when Tinker launched. Her framing wasn’t about building the smartest model. It was about making “frontier capabilities much more accessible to all people” — democratization as the mission, not capability supremacy. That is a genuinely different objective function than the one driving her former employer, and it was declared outright, not discovered after the fact to explain a disappointing benchmark result.

Inkling is a 975-billion-parameter mixture-of-experts model trained on forty-five trillion tokens across text, image, audio, and video, with a context window stretching to a million tokens. That is frontier-scale compute expenditure. This isn’t a company that ran out of runway and settled for a smaller ambition. It’s a company that spent frontier-level resources and then declined to spend the final increment chasing benchmark supremacy, presumably because the return on that increment doesn’t show up in the business they’re building.

A second detail: Inkling reportedly uses one-third the tokens of Nemotron 3 Ultra to hit equivalent performance on agentic coding benchmarks. That’s not a capability retreat — that’s a capability choice, optimizing for efficiency and cost-per-task rather than raw benchmark position. And the company is previewing a smaller sibling model alongside Inkling, suggesting a family strategy across sizes rather than a single mid-tier release.

The restaurant next door: Palantir

Thinking Machines isn’t the only company making this bet — and looking at who else is making it shows not everyone is occupying the same layer.

Earlier this month Palantir and Nvidia announced a “Sovereign AI Operating System” — Nvidia’s open Nemotron models, fine-tuned on a customer’s own data, running on Nvidia hardware inside that customer’s own air-gapped network, with Palantir’s Ontology and Foundry software layered on top. CEO Alex Karp pointed out that his enterprise customers don’t want to risk sharing their IP with frontier model providers and asked simply why wouldn’t they control the weights?

It’s tempting to read this as the same argument Thinking Machines is making. It isn’t, quite. What Palantir is selling is the restaurant: the finished, seasoned, plated product — an air-gapped AI system wired into a specific government agency’s or enterprise customer’s actual workflows, with “you control the weights” as the pitch that closes the deal. Palantir isn’t growing wheat. It’s the chef, working with ingredients somebody else grew. Somebody who could be trusted.

Another restaurant: Sierra

Sierra, Bret Taylor and Clay Bavor’s customer-support agent company, makes the same choice even more starkly. Sierra’s own technical writing describes a “constellation of models” architecture: rather than betting on a single LLM, Sierra routes each task inside a customer-service agent to whichever model — from OpenAI, Anthropic, Meta, or elsewhere — handles it best, and explicitly says it invests “in fine-tuned models where off-the-shelf models fail to meet our constraints.” Fine-tuning shows up in Sierra’s stack as one tool among several, alongside retrieval and layered “supervisor” models that catch mistakes before a customer sees them. Sierra has no interest in being a model company or an infrastructure company. It wants to be the restaurant that happens to keep a few specialty ingredients in the walk-in that nobody else stocks, because the dish needs them and they know just how to include them.

Mapping the rest of the stack

The farm-kitchen-restaurant split shows up everywhere the fine-tuning economy has organized itself.

The kitchen-builders — companies selling fine-tuning infrastructure to whoever wants to cook with it, indifferent to what gets made — now form a crowded field: Thinking Machines’ Tinker, Together AI, Fireworks AI, Predibase, OpenPipe, Baseten, Modal, Databricks’ Mosaic stack, and newer entrants like Nebius’s Token Factory and Prime Intellect. None of them care whether you’re building a coding agent, a legal research tool, or a customer-service bot.

The restaurants — companies where fine-tuning is invisible plumbing inside a finished, vertical product — include Palantir and Sierra, and many others. The addressable market for fine-tuning seems to include almost every possible enterprise adopting AI.

What’s seems unusual about Thinking Machines is that it’s trying to be the farm and the kitchen simultaneously while declining, by its own public admission, to be the restaurant. Most companies pick one layer and defend it. Thinking Machines is betting that owning two of the three is the more durable position — grow the flour, own the stove, and let Palantir, Sierra, and a thousand enterprise engineering teams fight over who plates the dish.

The same stack, built by design

As I was thinking about this, I wondered how this relates to the AI activities underway in China. It seems that China’s AI industry maps onto this same three-layer structure with unusual clarity — and one genuine wrinkle the American version doesn’t have.

The farm is crowded and innovating on a different axis than size: DeepSeek, Alibaba’s Qwen, Zhipu AI, Moonshot AI, MiniMax, ByteDance’s Doubao and Seedance. The standout isn’t scale, it’s efficiency — DeepSeek’s V3.2 reportedly uses a novel sparse attention mechanism to nearly match GPT-5 and Gemini 3 on complex reasoning despite far less compute, a different kind of farming: not more wheat, but wheat bred to need less water. Qwen has become the default soil for the rest of the world’s kitchens, generating over 100,000 derivative fine-tunes on Hugging Face. VC’s in Silicon Valley note how frequently their startup companies are building on Qwen.

The kitchen layer has its own SiliconFlow — a Beijing infrastructure startup, backed by Alibaba Cloud, that bills itself as the neutral layer between AI applications and hardware. It solves a problem others never had to: China’s compute runs across fragmented domestic chips, Huawei’s Ascend line chief among them, that don’t share Nvidia’s CUDA ecosystem. SiliconFlow abstracts that fragmentation away — it became the fastest platform serving DeepSeek traffic, and the only large provider running DeepSeek on Ascend chips instead of Nvidia’s. That’s a stove engineered to burn whatever fuel is in the tank that week, a direct product of the U.S. chip export controls rather than any inherent technical edge. Volcano Engine, Alibaba Cloud’s PAI, and Baidu’s Qianfan are versions of the same layer.

The restaurant layer is where China’s picture diverges most from Palantir and Sierra’s venture-funded improvisation: it’s named industrial policy.

Beijing’s “AI+” initiative targets seventy percent sectoral AI penetration by 2027, ninety by 2030 — fine-tuned vertical deployment treated the way past five-year plans treated high-speed rail. The players read like a sector directory: SenseTime for vision and embodied AI, iFlytek for speech in education and government, Baichuan Intelligence for healthcare, 4Paradigm for finance and industry, each fine-tuning a general base into something that only makes sense inside one workflow — a hospital’s diagnostic support tool, a bank’s risk model, an industrial inspection line.

The bet

Every frontier lab is implicitly betting that intelligence is the scarce resource, and that whoever has the most of it wins the enterprise market by default. Thinking Machines, Palantir, Sierra, and many others are all, in their different ways, betting against that premise — that raw intelligence is commoditizing faster than the frontier labs’ spending would suggest, and that the scarce resource has already migrated to whichever layer turns a generalist model into a specific customer’s model.

Thinking Machines is betting the moat moved to the farm-and-kitchen layer. Palantir, Sierra and others are betting it moved further still, to the restaurant, where nobody cares whose flour was used as long as the dish is right. China is betting on all three layers at once, with the state underwriting the bet directly.

It is a curious thing for me to watch companies with this much money and this much talent choose not to fight for the title of smartest model in the room. It is also a curious thing to watch them explain why, in public, in the first paragraph of an announcement.

But I think I’m beginning to understand.

Categories
AI China

Model Competition

The Chinese AI company Moonshot AI released its Kimi K3 model yesterday. I played with it a bit in the Kimi iOS app and, frankly, was underwhelmed. It was slow and disappointing. But this was hardly an extensive test. It’s strictly my first impression. I’m not doing coding. I’m what model makers would call a knowledge worker.

This morning Gavin Baker posted on X about Kimi K3 calling it potentially an “important inflection point for AI”:

Kimi K3 may be an important inflection point for AI. Potentially negative for Anthropic and OpenAI while being net positive for essentially every other company in the world. I mean that very literally. Although the real “Sputnik moment” would be an open-source frontier model that was also token efficient unlike Kimi K3 which is 50-70% more expensive to run than GPT 5.6 per Artificial Analysis.

Rationale:
 
A world where there are only 2-3 dominant frontier labs with 90% inference margins is net negative for every other layer while being awesome for those 2-3 labs. Those labs would become monopsonies for power, data centers, semiconductors and hyperscalers and would obviously vertically integrate over time into all those layers while also completely subsuming the application/software layers. 
 
Anything that lowers margins and increases competition at the model layer is good for every other AI layer: power, semiconductors, hyperscalers, neoclouds and yes even software.
 
This is why Jensen is so supportive of open-source. An open-source model requires the exact same amount of compute to run as a closed frontier model of similar size and architecture. Kimi K3 is roughly the same price as GPT 5.6 Terra on a per token basis, which actually suggests that it is less computationally efficient as I am sure that GPT 5.6 is priced to a higher margin than K3. And given that K3 is a token wastrel, i.e. token inefficient, it is significantly more expensive per task than GPT 5.6 and Grok 4.5, which are much more token efficient. Cost per token and token efficiency (i.e. intelligence density per token) are the drivers of intelligence per unit of cost. The winning AI companies will be those that offer the most intelligence per $ over time.
 
Lower margin % at the model layer = more margin $ at every part of the infrastructure layer and is a godsend for software. This can happen either through open-source models like K3 at the frontier or having a vertically integrated model company like Meta, SpaceX or Google at the frontier. Both outcomes result in a lower margin % at the model layer as vertically integrated model companies don’t really care where the margin $ come from. This is why it was so painful for OpenAI and Anthropic when Google was right there with them from a model competitiveness perspective and why Grok 4.5 and Muse 1.1 were just as important as Kimi K3.

The reason Kimi K3 is only potentially negative for Anthropic and OpenAI is 1) the @ericvishria point that the Claude and ChatGPT products and harnesses may be more important than their models today and 2) the hypothesis that they have much more advanced model checkpoints internally that are already being used for RSI. In the latter scenario, reaching RSI even a few months ahead of other labs might be enough to cement a permanent lead.

Time will tell on both points. And likely fairly quickly.

Caveat would be that since Kimi K3 is not token efficient and thereby actually more expensive than ChatGPT 5.6, we may need to see a more token efficient open-source model at the frontier or see Grok 5/Composer 4/Muse 2 at multiple points on the Pareto frontier for this potential risk to Anthropic and OpenAI to play out. And I am sure they will both vertically integrate as quickly as possible while continuing the product/harness strength they have shown over the last 8 months.

I wanted to explore this further so I asked Grok to explain his post:

This is a tweet (X post) from Gavin Baker (@GavinSBaker), Managing Partner and CIO of Atreides Management, posted on July 17, 2026.

It analyzes the recent launch of Kimi K3 (by Chinese AI company Moonshot AI) as a potentially significant shift in the AI industry. Baker argues it could hurt the high-margin dominance of closed U.S. frontier labs like OpenAI and Anthropic while benefiting nearly everyone else in the broader AI ecosystem (infrastructure, chips, data centers, power, hyperscalers, and software).

Quick Context on Kimi K3

Kimi K3 is Moonshot AI’s new flagship model (launched ~July 16, 2026). Key specs include:

  • ~2.8 trillion parameter Mixture-of-Experts (MoE) architecture.
  • Massive 1 million token context window.
  • Native vision capabilities.
  • Strong early performance in coding, agentic tasks, and long-context reasoning (competitive with or close to top closed models like GPT-5.6 variants or Claude models in certain benchmarks).
  • Pricing: $3 input / $15 output per million tokens (with caching discounts).
  • Important: Full weights are scheduled for open release (modified MIT license) around July 27, 2026 — making it one of the first truly open frontier-scale models.

It represents a step up from prior Kimi models and adds meaningful competition from a non-U.S. lab.1

The Two Images in the Tweet

These are screenshots from Artificial Analysis (an independent benchmarking site):

  1. Bar chart — “Cost per Intelligence Index Task” (lower is better). It breaks down costs by token type (input, output, reasoning, cache, etc.) and shows Kimi K3 as relatively expensive compared to some GPT-5.6 variants.
  2. Scatter plot — Intelligence Index (y-axis) vs. Cost per task (x-axis, log scale). It positions models in quadrants, with the “most attractive” area being high intelligence at low cost. Kimi K3 appears competitive on intelligence but not the cheapest.

These visuals support Baker’s point that while Kimi K3 reaches frontier-level intelligence, it is not the most token-efficient, making its effective cost per useful output higher than more efficient models.

Baker’s Core Argument (Broken Down)

Baker’s thesis is economic and structural, not just technical:

  • A world with only 2–3 dominant closed frontier labs (with ~90% inference margins) is bad for almost everyone else.
    Those labs would act as monopsonies (dominant buyers) for power, chips, data centers, and compute. They would vertically integrate upward (into infrastructure) and downward (into apps/software), capturing most of the value. Competition at the model layer prevents this concentration.
  • Anything that increases competition or lowers margins at the model layer is broadly positive.
    It redistributes economic value to the rest of the stack: more demand for GPUs/chips (inference especially), power, data centers, hyperscalers/neoclouds, and software (because cheaper/better AI drives wider adoption).
  • Token efficiency matters more than raw price per token.
    Baker notes Kimi K3 is roughly the same per-token price as GPT-5.6 but 50-70% more expensive overall according to Artificial Analysis. It is a “token wastrel” — it uses more tokens to achieve similar results (lower “intelligence density per token”).
    → True winners deliver the most intelligence per dollar. Cost-per-token + efficiency together determine real value.
  • Why NVIDIA’s Jensen Huang loves open-source models.
    Open models still require massive compute to run at scale (same as closed ones). More models + wider adoption = more inference demand = more GPUs sold. A few ultra-high-margin closed labs would concentrate spend and reduce overall volume.
  • Vertical integration changes the game.
    Companies like Google, Meta, or xAI/SpaceX that have both strong models and their own infrastructure don’t obsess over model-layer margins — they capture value elsewhere. This is why more competition (open or vertically integrated) hurts pure-play high-margin model companies like OpenAI/Anthropic.
Why It’s Only “Potentially” Negative for OpenAI and Anthropic

Baker is careful:

  • Their products, user interfaces, and “harnesses” (the surrounding tools/ecosystems) may matter more than raw model performance right now.
  • They likely have more advanced internal models/checkpoints (possibly already used for recursive self-improvement/RSI).
  • Kimi K3’s token inefficiency softens the blow for now. A more efficient open-source frontier model (or stronger showings from Grok, etc.) would be more threatening.
  • OpenAI and Anthropic are already moving to vertically integrate.
Overall Takeaway from the Post

Baker sees Kimi K3 (and the wave of strong Chinese models like Qwen, DeepSeek, GLM) as healthy competition that prevents unhealthy concentration. It’s not yet a full “Sputnik moment” (a truly dominant, efficient open-source frontier model), but it points in that direction.

Net effect he describes: Lower % margins at the model layer → more total margin dollars flowing to infrastructure, chips, power, and software layers. This is bullish for the broader AI ecosystem even if it pressures the pure closed frontier labs.

The tone is analytical and balanced — he acknowledges uncertainties and that time will tell (likely quickly). Replies to the post generally agree this kind of competition is positive for the industry overall and could shift more spend toward efficient inference hardware.

This fits Baker’s broader investing views on AI: he often emphasizes infrastructure winners, token economics, and how value accrues across the full stack rather than just at the smartest model.

Interesting.

Categories
AI

Memory Optimization

After thinking about the very high margins that Micron recently reported (greater than Nvidia’s) I asked Grok 4.5:

What startups are pursuing technology that reduces memory requirements for ai models and inference?

Several startups are actively developing technologies to slash memory requirements for AI models and inference. These efforts target the core bottlenecks in large language models (LLMs) and multimodal systems—primarily through extreme quantization (reducing bit precision of weights), alternative architectures that avoid quadratic memory scaling (e.g., state-space models or hybrids), sparsity/partial activation (like Mixture-of-Experts or neuron-level sparsity), efficient runtimes/kernels, and hardware innovations like in-memory computing that minimize data movement.

This space is especially active for on-device/edge AI (phones, laptops, embedded systems) and cost/energy-efficient cloud inference, where fitting larger or more capable models into limited RAM/VRAM, reducing KV cache bloat in Transformers, or lowering bandwidth demands delivers major gains.

Here are some of the most prominent startups and their approaches (as of mid-2026):

Extreme Quantization & Model Compression

  • PrismML (Caltech spinout, backed by Khosla Ventures): Focuses on extreme low-bit architectures—1-bit binary (“Bonsai”) and ternary (1.58-bit, weights in {-1, 0, +1}). They compressed Alibaba’s Qwen3.6 27B model from ~54 GB (FP16) to under 4 GB and ran the full dense 27B model on an iPhone 17 Pro. Claims include up to 14× smaller memory footprint, 8× faster inference, and significantly lower energy use, with competitive or better benchmark performance. They have open-sourced Bonsai models (including smaller 8B/4B/1.7B variants) under Apache 2.0 and are in discussions with Apple. This represents one of the most aggressive commercial pushes into 1-bit/ternary models for on-device deployment.
  • Mobius Labs (Berlin): Developed Half-Quadratic Quantization (HQQ), a fast, calibration-light post-training quantization method that enables high-accuracy low-bit models (including aggressive 2-4 bit). They demonstrated quantizing Llama 70B to run on a single GPU instead of four without major accuracy loss, directly cutting memory and compute needs. Their work extends to FP4 optimizations and integrates with frameworks like vLLM.
Alternative Architectures for Inherent Memory Efficiency
  • Liquid AI (MIT spinoff): Builds Liquid Foundation Models (LFM / LFM2 series)—hybrid architectures combining gated short convolutions with grouped-query attention (GQA) blocks, plus MoE variants. These deliver substantially lower memory footprints than Transformers (especially for long contexts, avoiding massive KV cache growth), faster prefill/decode (up to 2× on CPU in some cases), and strong on-device performance. Examples include tiny models (230M–350M params, often
  • Cartesia: Specializes in state-space models (SSMs) and Mamba-derived/hybrid architectures. These offer near-linear (or better) scaling with sequence length and near-constant memory complexity during inference (fixed or minimal growth vs. quadratic KV cache in Transformers). This enables efficient long-context processing on smaller devices with lower energy use. They provide on-device libraries (e.g., “Edge”) and models like hybrid Mamba-2 variants optimized for throughput and fixed memory footprints.
On-Device/Edge Runtimes & Optimizations
  • Cactus (Y Combinator-backed): Builds cross-platform, energy-efficient inference runtimes and kernels for mobile, wearables, and low-power devices. Supports aggressive quantization (down to 2-bit), delivers very low latency (sub-50 ms time-to-first-token), and keeps models small (e.g., hundreds of MB for small models). Emphasizes privacy (fully local) and broad model support (Llama, Qwen, Gemma, etc.).
  • Embedl: Focuses on edge-optimized multimodal models and mixed-precision quantization (e.g., W4A16-Edge2 with selective FP16 layers + “FlashHead” optimizations). Enables fast multimodal reasoning on devices with under 8 GB RAM (e.g., Jetson Orin Nano/AGX), with significant speedups over baselines while recovering most accuracy.
Hardware Innovations Reducing Memory Pressure
  • d-Matrix: Develops digital in-memory computing (DIMC / 3DIMC) chips and platforms (e.g., Corsair) that integrate compute directly with/near memory (initially SRAM, now advancing to 3D DRAM). This drastically cuts data movement—the dominant memory bandwidth and energy bottleneck in inference. Claims include major speedups (up to 10× in some workloads) and better efficiency vs. traditional GPU/HBM setups, enabling higher performance with more manageable overall memory system requirements.
  • Hailo: Designs efficient edge AI inference chips (e.g., Hailo-10H series) optimized for low-power LLM and multimodal workloads. Their chips target 1–3B parameter models at very low power envelopes (~2–2.5 W), making deployment feasible on memory- and power-constrained edge devices where traditional hardware would struggle.
Other Notable Mentions
  • Tiiny AI: Uses neuron-level sparse activation (“TurboSparse”) + heterogeneous engines for running very large models (up to 120B) fully on-device/offline in compact hardware.
  • Broader ecosystem players (inference platforms like Together AI or Fireworks) heavily optimize serving (quantization, caching, KV cache management) but are more infrastructure-focused than core technology innovators in compression/architecture.
Key Trends and Context

Quantization (especially post-training or quantization-aware) remains the most immediate lever and is widely adopted. Architectural shifts (SSMs, hybrids, MoE with low active parameters, convolutions) address fundamental scaling issues like KV cache memory growth. Hardware approaches (in-memory compute, high on-chip SRAM, efficient NPUs) tackle bandwidth and data movement. Many solutions combine these (e.g., architecture + quantization + sparsity).

Progress is rapid, driven by on-device AI demands (Apple’s interest in PrismML is notable) and the economics of scaling inference. Many models are open-sourced or have open weights, accelerating adoption. Results vary by workload—extreme 1-bit models excel in certain scenarios but may need custom kernels; SSM/hybrid models shine for long context or linear scaling needs.

For the absolute latest benchmarks or specific model deployments, check company sites (e.g., PrismML’s Bonsai releases, Liquid AI’s LFM models on Hugging Face) or recent evaluations, as the field evolves quickly.

Interesting!

Categories
Science

The Whiteness of the Lather

“Consider a bar of soap. Has it ever struck you that soap lather is always white no matter what color the soap is? That isn’t because the soap somehow changes color when it is moistened and rubbed. Molecularly, it’s exactly as it was before. It’s just that the foam reflects light in a different way. You get the same effect with crashing waves on a beach—greeny-blue water, white foam—and lots of other phenomena. That is because color isn’t a fixed reality but a perception.” (Bill Bryson, The Body)

I have been thinking about soap. Not the smell of it, not the brand, not the dish it sits in by the sink, but its color, and what becomes of that color the moment the soap is wet and worked between the hands. However dark the bar the foam that comes off it is white. It is always white.

The soap has not changed. This is the part that unsettles me a little, still, though I understand the physics of it well enough to explain it to a child. Nothing has been dissolved out of the soap, no pigment washed down the drain with the first rinse. The bar, examined afterward, is exactly what it was — same weight, same tint, same faint milled scent. What has changed is only the shape of the surface presented to the light. A solid surface reflects color; a foam of small bubbles scatters light in every direction at once, and the scattering reads, to the eye, as white. The water does this too, at the shoreline, where the sea that looked green or the color of a bruise all the way to the horizon turns to white foam the instant it breaks against the sand. Nothing in the ocean has been added or subtracted. Only its surface has been shattered into a million small mirrors, and a million small mirrors, taken together, show us nothing but the light itself.

There is a machine built on the same principle, aimed at the opposite end. A stealth aircraft is not invisible in the way a child means invisible — it has not been made transparent, or absent, or small. It has been shaped. Every facet of the old F-117, every curved plate of the B-2, is angled to take an incoming radar signal and send it away, off at some oblique angle, into empty sky, anywhere but back to the dish that sent it out looking. Where the geometry alone cannot manage it, a coating finishes the job, converting the beam into a small, undetectable warmth rather than an echo. The aircraft has not changed what it is. It has changed only where its reflection goes. The operator at the radar screen sees nothing, and concludes, wrongly, that there is nothing there.

The foam and the aircraft are the same trick, pointed in opposite directions. Both break a surface’s geometry until the return signal — light, in one case, radar, in the other — stops describing the object honestly. The foam scatters light everywhere at once and reads, to the eye, as brilliant and total. The aircraft scatters radar everywhere except back to the observer and reads, to the screen, as nothing at all. One hides by flooding the return. The other hides by starving it. Neither the soap nor the aircraft has changed what it is. What has changed is the transaction between the object and whatever signal goes out to find it — and it is that transaction, not the object, that a color or a presence turns out to describe.

I do not think this makes color, or presence, less real. I think it makes real a smaller and stranger category than we had assumed. We want our perceptions to be reports from the world, filed accurately, awaiting only our attention. It is more honest, and harder to sit with, to admit that a great many of our perceptions are not reports at all but events — things that happen at the meeting point of a self and a surface, and that would not happen in quite the same way to anyone standing an inch to the left.

I rinse the soap and set it back in the dish. In a moment the foam is gone, down through the drain, and the bar is only itself again, the color it always was, waiting for the next hand to come along and change it into something briefly, purely white.

Categories
AI Business

The Reverse Information Paradox We’ve Always Had

Satya Nadella wrote recently about what he calls the Reverse Information Paradox: enterprises pay for AI intelligence twice. Once in money. Again in the proprietary knowledge they surrender through every prompt, correction, and evaluation. The better they use the model, the more of their own institutional understanding leaks into someone else’s system. The vendor ends up knowing more about the buyer’s business than the buyer knows about what the vendor retained.

Replace “model” with “employee” (or “consultant”) and the paradox is not new at all.

You pay for a person once with salary. You pay again with something harder to price: the context, relationships, and judgment they must absorb to become useful to you. The better they perform, the deeper the immersion, the more of your particular way of doing things moves into their head. Every correction and late-night conversation is another trace of institutional memory changing hands. When they leave, some of that memory leaves with them. Not always through theft. Usually just through the ordinary residue of good work.

The visible cost is salary; the invisible cost is the slow transfer of what makes you distinctive. High performers get more access precisely because they’re high performers, which means the leakage accelerates exactly when you can least afford it. The exhaust is just harder to see with people than with tokens — it moves through conversation and mental models instead of logs.

The analogy has a limit, and the limit matters. Employees bring knowledge in, not just absorb it. They have judgment and relationships a model doesn’t. Models are purely absorptive, and once something is inside them, it’s infinitely reproducible — a person can only be in one place, working for one employer, at a time. We’ve had a few hundred years to build tools for the human version of this problem: contracts, culture, non-competes. The model equivalent is still being invented in real time, which is exactly why Nadella felt the need to name it.

Apple’s recent legal action against former employees who joined OpenAI is this pattern in its sharpest form. Whatever the specifics, the shape is familiar: people who spent years inside one of the most sophisticated organizations in the world, carrying out knowledge that never appeared on any balance sheet and was hard to contain. No one fully anticipates what a mind absorbs simply by being in the room long enough.

That’s the real difference between the silicon case and the human one. You can try to take action to wall off knowledge flowing to a model. You cannot wall off what someone has learned to notice.