The Sorting Machine: Everybody's working. Nobody's moving.

Happy Sunday.

Do one thing for me before you read the rest.

Picture someone you know between 22 and 45. Out of school, somewhere in the thick of building a career still. Got a face? Good. Hold onto them, because by the end of this I think you're going to want to forward it their way.

Two numbers from last week are rattling around my head. On Friday the government released a strong jobs report. 172,000 new jobs, more than double what economists expected, the third month running of beating the forecast, unemployment holding at 4.3 percent. Basically, the read writes itself: the AI scare has been overblown, the economy's fine, go enjoy your weekend.

But…. then,, in the same week a second number came in. The firm that tracks layoffs said AI was the number one reason American companies gave for cutting jobs in May. Forty percent of the month's cuts. More blamed on AI in the first five months of this year than in all of last year combined.

So which is it. Booming, or bleeding?

Neither, I think. The real theme is hiding underneath both of them.

The market isn't frozen. It's sorting.

That 4.3 percent is lying to you a little. The rate is calm, but not because companies are hiring like crazy. It's calm because almost nobody is getting fired either. Economists have a boring name for it, low-hire, low-fire, and it means just what it sounds like. If you already have a seat, you're fine. If you're trying to get one, or trade up for a better one, the door is heavier than the headline makes it look. And the people who do lose a job right now wait longer than they have in years to find the next.

You can see it in a single number. Plenty of jobs are posted, but companies are filling fewer of them. That's not a business that stopped hiring. That's a business with a line in the water, watching what swims by and reeling in almost nothing. Picky, not paralyzed.

And every one of those jobs is quietly running through the same test: keep this person, replace this person, or turn them into the one who aims the thing.

Nobody likes to say who that test is built to reward. Everyone assumes it's the generation that grew up fluent in the machine. I think that's the most expensive wrong guess in this whole conversation.

The most prepared generation is the one struggling

New grads are having a brutal time finding work. That part's real. Unemployment for college grads aged 22 to 27 is sitting near 5.6 percent, about the worst it's been outside the pandemic, and roughly nine in ten of this year's grads say they're scared AI will take the entry-level job before they can land it.

Now sit with the odd part. This is the most AI-native group of humans we've ever produced. They can drive these tools in their sleep. So why are they the ones stuck on the outside?

The tools that made them capable made them identical.

When everyone's cover letter is polished and everyone's application is sharp, none of them stand out. So a recruiter facing four hundred flawless résumés stops reading and falls back on the one thing a machine can't fake. Who do I already know? Who got vouched for? The great equalizer made who-you-know worth more, not less.

Peter Cappelli, a management professor at Wharton, puts it bluntly: there's almost no skill in using one of these tools. If you can run a Google search, you can run one of these. The fluency everyone brags about isn't an edge. It's the floor. Everyone stands on it now, which is exactly why it's worth close to nothing on its own.

So if the tool does the task, and everyone has the tool, what's left worth paying for? Knowing what to point it at. Which turns out to be a completely different muscle.

Today’s Sponsor

The Great Wealth Transfer: What Happens When $124 Trillion Changes Hands?

New projections earlier this year have the Great Wealth Transfer at a potential $124 trillion through 2048.

BofA has been tracking younger high net worth investors closely. 83% of those 43 and younger said they currently own art or aspire to. This is compared to just 34% of those older than 43.

Meanwhile, the number of Warhols and Picassos isn’t going up.

Masterworks lets you invest in the art market before the full weight of that decades-long wealth transfer hits. 

70,000+ investors have already deployed $1.3 billion across 500+ artworks by Banksy, Basquiat, and many more. 29 sales to-date have delivered net annualized returns like 16.5%, 17.6%, and 17.8% on those held longer than one year, not including those unsold.

This is your opportunity to invest in the art market. Click here to skip the waitlist.

^{Past performance is not indicative of future performance. Investing involves risk. See important disclosures at masterworks.com/cd}

The people who price "valuable" for a living already ran this experiment

You don't have to take my word for it. Take it from the people whose whole job is figuring out what's worth something: hedge funds.

I've been chewing on a talk by a fund manager named Alix Pasquet, who runs Prime Macaya. The talk is about why smart people lose money, and he lays out a problem that's basically this one. Funds don't really want to hire junior analysts anymore, for two reasons stacked right on top of each other.

One: the AI now does the junior work. The modeling, the data pulls, the grunt a 23-year-old used to cut their teeth on. Done, faster and cheaper.

Two: the junior still can't do the senior work. Ask a young analyst what the CEO is really thinking behind the careful words on the earnings call, and they freeze. They can build you a perfect model. They can't tell you that this quarter rhymes with a bunch of stocks that blew up in 1999, or make the jump from "sales look soft" to "the real problem is the board." That's judgment. Reading between the lines. Spotting the pattern. Making the leap.

Where does that come from? Reps. Slow ones. Pasquet has a name for it I keep coming back to: analog training. Sitting with something hard, under real limits, long enough to build a picture in your head that nobody can hand you. His ideal is almost quaint. A company's annual report and a pencil. Get good enough that you don't need the tool, and then the tool makes you dangerous.

Read that one twice. The machine doesn't reward the people fluent in the machine. It rewards the people with the judgment to aim it. Fluency is the cheap part. Judgment is the scarce part. And it's the scarce part the youngest workers got the fewest reps to build.

Picture it on the ground. For sixty years, Phoenix ran on customer-service floors and back-office desks. Easy to wave those off as low-skill jobs. Wrong read. Those rooms were one of the biggest unofficial schools in the country, where a kid with no degree learned to untangle a stranger's problem, take a beating on the phone, and climb. People started on a call-center line and ended up running things. That's the rung getting sawed off first. We didn't just cut some jobs. We cut the bottom off the ladder.

Now the honest other side

Two things keep this from being a doom story.

First, that scary AI-layoff number is weaker than it looks. The firm that publishes it just tallies what companies tell it. And "we cut jobs because of AI" is a very convenient thing to say. It sounds visionary instead of "we over-hired in 2021 after laying everyone off during covid." Even Sam Altman, who runs OpenAI and has every reason to talk AI up, called it out. He says companies are AI-washing their layoffs, blaming the robot for ordinary cost-cutting. Glassdoor's economist agrees, and Apollo's chief economist goes further: he sees zero hard proof of jobs actually lost to AI. Treat the number as a real signal, not a verdict.

Second, this movie has played before. Big leaps in productivity have a long record of destroying some jobs and creating even more, just later, and usually somewhere nobody was looking. Cappelli's other point: wiring AI into a real company is slow, expensive, and clunky, far harder than the press releases suggest. The wipeout everyone braced for hasn't shown up on time.

So this isn't the sky falling. It's a sort. And the good thing about a sort is that you can study what it rewards and go become it.

What do you do about it?

Strip it all the way down and you get something almost annoyingly simple.

The market is paying up for one thing, and it isn't AI fluency. Everyone has that, and alone it's worth about as much as knowing how to work a search bar. It's paying for judgment. Reading the room behind the words, seeing the pattern, making the leap, holding the whole picture in your head and aiming the machine at the part that matters.

And the good news, the part that should feel like a door opening instead of closing: judgment can be built. It always could. It just gets built the slow, unglamorous way. Read the actual book, not the summary of the summary. Stay with a hard problem past the point of boredom, because the real insight usually waits on the far side of bored. Get the reps. Find someone older who's been through a few cycles and steal their scars, because right now that handoff is the most underpriced trade going.

If you've got money in the market, there's a version of this for you too. Next time a company brags that AI is its moat, raise an eyebrow. Everyone has the same tools. The tool was never the edge. The edge is knowing where to aim it, and that's the part nobody can copy.

So go back to the person you pictured at the top. Maybe it's the grad firing two hundred applications into the void. Maybe it's your kid. Maybe it's you. The question worth asking isn't whether they can use the tools. They can, probably better than you or me.

The real one is this: did anyone ever teach them how to think? How to aim the thing at something real?

Go text them.

 Stay curious 😎

- John

Today’s Sponsor

Exclusive Briefing: SpaceX IPO 2026

Most retail investors will hear about this after it's too late. Discover the verified signals Wall Street is watching and the access paths most people don't know exist.

Download now