They're Not Laying You Off. They're Just Not Replacing You.

JPMorgan is eliminating 30,000 jobs a year. No press releases. No headlines. Just attrition — and silence where those seats used to be.

In a recent CNBC interview, Jamie Dimon said it plainly. JPMorgan runs 10% annual attrition across its 300,000-person workforce. He's not backfilling those seats. Every app, every process, every job will be affected. And when the interviewer pushed him on whether this was just lower-level roles — Dimon said no. "I think it'll be more than you think."

That's the CEO of the most powerful bank in America, on camera, telling you the restructuring is already happening. Not in five years. Now.

The question isn't whether this is coming. Dimon just confirmed it is. The question is: what profile survives it — and what profile gets quietly erased?

How Attrition Became the New Layoff

There's a reason you're not reading about 30,000 JPMorgan job cuts in the news. They're not job cuts. They're just... gaps that never get filled.

Someone retires. Someone leaves for another company. Someone burns out and walks away. In the old playbook, HR posts the role, screens candidates, brings in someone new. The headcount stays flat. The work continues.

That's not what's happening now.

What's happening now is the role gets handed to an AI workflow, or to a team that's running smarter with better tools, or to a new hire who's half the cost and twice the output because they're operating with AI as a co-pilot. The seat disappears. Nobody announces it. Nobody protests it. The org chart just quietly gets a little shorter.

This is how large institutions restructure without triggering the optics of a mass layoff. It's cleaner. It's slower. And it's virtually invisible — until one day you look around and realize the entire floor looks different.

Dimon was refreshingly honest about the mechanism: "We have 10% attrition a year, which means our headcount is going down 25, 30,000 a year. And we are going to be prepared to say, okay, we love these people... We're going to give them reskilling, new skills, better jobs, move them somewhere else, maybe early retirement."

Read that carefully. He's not describing a workforce reduction. He's describing a workforce transformation — one that happens quietly, role by role, through the natural churn of people leaving. The company doesn't shrink. It reshapes. And if your role is in the part being reshaped, you may not find out until you're the one leaving and nobody's looking to fill your position.

"It Won't Just Be Lower-Level Roles"

Here's the part that should cut through the noise for anyone reading this who has spent 20 or 30 years building a career in a corporate environment.

There's a common assumption among senior professionals that AI automation is a threat to entry-level and mid-level workers — the data processors, the analysts, the coordinators, the report writers. People who do repetitive, structured work that's easy to automate.

That assumption is wrong.

When Dimon was asked directly whether this was primarily a lower-level job issue, he said no. He said it'll affect every level. He said it'll be "more than you think."

Senior roles are changing too — just differently. The VP who used to manage a team of 12 analysts who each ran one model now manages two people and a suite of AI tools that do what the team of 12 used to do. The Chief Risk Officer who used to rely on a department of specialists to stress-test scenarios now has AI systems that run a thousand scenarios overnight. The work that justified large, expensive teams is getting compressed into smaller, smarter, AI-augmented configurations.

That doesn't mean the senior leader disappears. It means the senior leader without AI fluency disappears — and gets replaced by one who has it.

The Profile That Gets Hired (Not Eliminated)

So who IS JPMorgan hiring to fill those new AI specialist seats? It's worth thinking carefully about this.

They're not hiring people with no domain knowledge who happen to know how to use AI tools. A 28-year-old who's great at prompt engineering but has never worked in financial services, risk management, or lending operations isn't going to run point on transforming JPMorgan's credit underwriting workflow. They don't know enough about what can go wrong. They don't know the regulatory environment. They don't know where the AI will confidently produce output that's technically coherent but operationally wrong.

They're also not hiring senior professionals who have deep domain expertise but can't operate in an AI-augmented environment. The 55-year-old risk executive who dismisses AI tools as unreliable or refuses to learn how to work with them is getting managed out — quietly, through attrition — just as surely as the analyst whose job got automated.

The profile that survives — the profile that gets hired, not replaced — sits at the intersection of both. Deep domain expertise plus AI fluency. That's the combination.

I call these people domain translators.

A domain translator is someone who has spent 15 to 25 years inside a specific industry or function — finance, healthcare, supply chain, cybersecurity, engineering, HR, whatever — and has developed the ability to deploy AI inside that domain in ways that produce real business value. They understand the domain deeply enough to know what good output looks like, where the risks are, and what questions to ask. And they're fluent enough with AI to actually build and direct the workflows.

They're not AI engineers. They're not prompt engineers. They're experienced professionals who've added a critical second skill set.

Why Your 20 Years Is the Hard Part

Here's the thing most people miss when they hear this framing: if you've been working in your field for two decades, you already have the hard part.

The domain knowledge isn't something you can shortcut. You can't learn 20 years of pattern recognition in six months. You can't replicate the institutional knowledge, the industry relationships, the hard-won intuition about where projects go sideways and why. That's what made you worth $200,000, $300,000, $400,000 a year. That expertise is real, and it doesn't evaporate.

What's changing is the leverage on that expertise.

AI tools are multipliers. In the right hands — hands that understand the domain — they compress what used to take a team into what one experienced professional can do independently. That's not a threat to the experienced professional. That's an extraordinary upgrade in capability.

But only if you build the second half.

The professionals who are getting squeezed out aren't the ones with the least experience. They're the ones who stopped developing. Who assumed that deep domain knowledge was sufficient and that the tools were someone else's problem. Who watched AI arrive in their industry and waited to see how it shook out instead of getting ahead of it.

That's the real risk. Not that AI replaces you. But that it makes you obsolete by making someone else 10x more capable.

Two Paths Forward — Both Are Real

I work with senior professionals every week, and I see two distinct paths playing out right now. Both are legitimate. Neither is wrong.

Path One: Upgrade Your W-2 Profile

For many people in their 50s with strong networks in their industry, the best move is staying in the W-2 world — but upgrading the profile significantly. That means developing genuine fluency with AI tools relevant to your domain, building a visible body of work that demonstrates how you use them, and positioning yourself explicitly as a senior professional who bridges deep expertise and AI capability.

Companies are actively looking for people who can do this. They're not always advertising for it explicitly — the job posting might say "Director of Operations" or "VP of Finance" — but when they interview, they're evaluating whether candidates can operate in an AI-augmented environment. The ones who can demonstrate that command a premium. The ones who can't are increasingly invisible.

This path is about protecting and extending a W-2 career by being explicitly, visibly AI-fluent in your domain.

Path Two: Build an Independent Practice

The second path is building an advisory or fractional practice around your domain expertise plus AI fluency. This isn't for everyone, and it's not the right call in every situation. But for senior professionals who have genuinely distinctive domain knowledge and are willing to think differently about how they deliver value, it's a real option.

The model is straightforward: three to five companies, each paying a monthly retainer for a senior professional who handles a specific function part-time, augmented by AI tools that make the economics work. The AI isn't replacing the professional — it's making it possible for one professional to serve multiple clients at a level that used to require a full-time hire at each.

This path requires building a brand, developing an outreach strategy, and being willing to sell. It's not passive. But the upside is significant independence and the ability to monetize expertise across multiple relationships simultaneously.

Most smart professionals I know are building toward both tracks at the same time — because the window won't stay open forever.

The Window Dimon Just Confirmed Is Open

What Jamie Dimon confirmed in that CNBC interview isn't alarming. It's clarifying.

The restructuring is happening. The seats are disappearing. And the new seats — the ones being filled — have a specific profile requirement. Domain expertise plus AI fluency.

If you've spent 20 years building deep expertise in your field, you are not starting from zero. You're starting from a position of significant advantage over anyone who's trying to enter this space without that background.

But the advantage only holds if you close the loop. If you build the second half. If you stop treating AI fluency as someone else's job and start treating it as yours.

The professionals who figure this out now — who build the combination before the window closes — are going to look remarkably well-positioned 18 to 24 months from now. The ones who wait are going to be competing for roles that are increasingly scarce, against candidates who started building earlier.

Dimon put the number on the table: 30,000 seats per year at one company. Multiply that across every major financial institution, every large enterprise, every sector going through the same transformation.

The seats aren't disappearing. They're changing. The only question is whether your profile changes with them.

Ready to Figure Out Your Next Move?

• Schedule a Strategy Call
• Join My Newsletter