Dear Friends,
Last Monday, two economists released a study that anyone who has spent decades getting very good at something should read.
Bharat Chandar of Stanford and Bouke Klein Teeselink of King's College London looked at more than a billion job postings and 154 million employment records across 41 countries. At companies that adopted AI, senior employment rose 6.7% over five years. Junior employment fell 3%.
If you're in the back half of a long career, sit with that for a second. The technology everyone warned would replace you is, for now, making you more valuable.
In July I wrote about research pointing the other way: workers 55 and older in AI-exposed fields leaving their jobs more often, some before they were ready. Both findings hold. The same technology is pushing some experienced people out and pulling others further in.
What separates the two groups isn't age. It's what the work actually consists of.
AI is very good at tasks. It drafts, summarizes, researches, formats. It is far less reliable at knowing whether the draft is any good. That takes judgment, and judgment is what the market is paying for right now.
Here is the part that deserves more attention.
Judgment doesn't come from nowhere. You built yours doing the work AI now does. The first drafts that came back covered in red. The analysis you got wrong in front of someone senior. The client call you fumbled and never fumbled that way again.
Researchers at Seoul National University recently interviewed senior developers and students about this and gave it a name: productive struggle. Their sample was small, but the idea will be familiar to anyone who learned a craft the long way. You don't get the judgment without the struggle.
Which means the ladder you climbed is being taken apart beneath you. A separate Stanford analysis found that employment for 22- to 25-year-olds in the most AI-exposed jobs is running 19% below where it otherwise would be. Last week I wrote about law firms whose senior partners won't retire, and the younger lawyers leaving because there's no room to lead. Now there are fewer junior seats to begin with.
So for the accomplished people I work with, this moment carries two messages at once. You are more valuable. And the thing that makes you valuable is becoming rare.
A senior executive I worked with recently put it this way: "AI made me more valuable this year, and I've never been less sure what I'm for."
That's the identity question hiding inside the labor data.
For most of a career, you know your value by what you produce. The deliverable, the deal, the result. When a tool can produce much of that faster, the value moves somewhere harder to see. It moves into what you notice. What you'd flag before anyone else does. The call you'd make when the data is ambiguous. The sense, earned over thirty years, that something is off.
Most accomplished people have never had to describe that. They've simply used it.
I think that's worth fixing, whatever you decide about what comes next.
If you stay, you'll need language for your judgment in an organization that is re-sorting itself around it. If you go, it's the most portable thing you own, and it's hard to carry something you can't describe. And if you're wondering what could make the next chapter meaningful, it may matter that what you have is something fewer people will get the chance to build the way you did.
I'm not suggesting everyone become a mentor. Some of you would hate it, and that's a legitimate answer. Knowing where you stand comes before deciding what to do with it.
So start there. Try describing your judgment to someone who has never done your job.
If you can't yet, that isn't a failure. It's the work.
Warmly,
Cara
Future Identity Strategist
founder, Good Morning Freedom
P.S. The Future Identity Snapshot looks at what you'd carry forward from the role you're in now, including the parts you've never had to put into words. Ten honest minutes: https://www.goodmorningfreedom.com/future-identity-snapshot
The executive described here is a composite, with details changed to protect privacy.