You’re Not Getting Replaced. You’re Getting Repriced
The scariest useful thing I’ve done with AI this year.
TL;DR: A viral AI doom essay says the economy could crack by 2028. Most people read it, shared it, and felt worse. I read it, opened Claude, and turned the scenario into a career strategy session. That difference, consuming fear vs converting it into a plan, is the repricing. This is what I built, what it showed me, and what you should do next.
It was 11:00 p.m. last night and I was in bed doing the exact thing I tell people not to do, (Doomscrolling) and the essay on my screen was making an annoyingly coherent case that the economy I’ve spent my entire career inside of might not exist in 30 months.
The piece was “The 2028 Global Intelligence Crisis.” Six thousand words with crystal clear logic.
Here’s what it said; Each domino after domino: AI automates knowledge work, margins collapse, white-collar income evaporates, and suddenly the mortgage market, thirteen trillion dollars of “this is fine”, starts asking questions that make everyone in finance sit up straighter.
I read it.
Then I did that thing where you put your phone down and stare at the ceiling because you can’t tell whether you just read something important or something engineered to hijack your nervous system.
And the worst part is, it can be both.
Then I did the most useful thing I’ve done with AI all year.
I didn’t share it with a fire emoji or post a hot take.
I didn’t argue in the comments, which took genuine restraint because I am absolutely the kind of person who argues in the comments….sometimes, don’t judge.
I opened Claude, and instead of asking spectator questions like “summarize this” or “should I be worried?” I built a prompt.
Not a prompt-prompt.
More like a brief.
The output would be something more like a document you’d hand to a serious analyst if you had one on retainer at midnight.
I don’t have one of those.
I have 3 kids 6 and under, a mortgage, and an AI subscription that costs less than Spotify. (Just a few Ai subscriptions)
So I made the tool do analyst work.
Here’s the core instruction set (lightly cleaned up so you can steal it):
ROLE: You are a cross-disciplinary strategy analyst (macro, labor markets, product strategy, risk).
TASK: Evaluate the essay’s scenario as a set of mechanisms, not a vibe.
OUTPUT:
What’s plausible vs what’s narrative glue
What breaks first, second, third (and why)
The non-obvious opportunities created by the disruption
“Repricing map”: which skills/roles get cheaper, which get more expensive
A plan tailored to my skill set (be brutally honest)
Then I added one line that doom content almost never includes:
“Now find the opportunities.”
Not feel-good optimism….I wanted specific positions that become either (a) hard to automate or (b) more valuable because automation is happening.
And because I’m apparently committed to emotional damage at inappropriate hours, I asked it to evaluate my own career viability in that future.
Asking a machine to grade your life at midnight while your spouse sleeps next to you is either a power move or a cry for help, still unclear.
The output was long and oddly/uncomfortably specific.
And it changed how I think about the entire genre of AI doom essays.
Because the essay modeled the technology beautifully.
What it didn’t model at all was humans.
Every doom scenario does the same lazy thing: it treats people like fixed job titles with no adaptation.
If your current work is disrupted, you drop straight to zero with NO reinvention.
No “22-year veteran decides on a random Tuesday to learn something new and actually does it.”
My prompt forced the model past that assumption.
And what came back was a version of the future the doom essay literally cannot see, because it’s not built to see it.
The actual pattern: destruction, survival, and the third bucket
When you stop asking, “Is this scary?” and start asking, “Where does value move?” the economy splits into three buckets:
Things that get commoditized
Things that survive mostly intact
Things that become more valuable because of the disruption
That third bucket is where your career lives or dies.
Here’s the pattern that kept showing up:
Most exposed (near-term)
Work that’s mostly information processing + standardized output.
routine analysis and reporting
generic content and research
mid-market advisory work where the “magic” is formatting information
roles that exist mainly as friction between two parties (This may be one of the most important lines in this post. If your work is predicated on removing friction it’s the most at risk for AI to replace)
Job titles you’ll feel this first:
Junior analyst / reporting analyst
Customer support manager (tier 1–2 ops)
SEO/content strategist at a commodity agency
Sales development rep (high-volume outbound)
Contract reviewer / paralegal (routine work)
Least exposed (durable)
Work where being human isn’t a feature, it’s the product.
hands-on care (health, elder care, childcare)
high-trust, high-stakes human moments (grief, crisis, trauma, conflict)
physical presence and skilled trades
leadership that calms people down and coordinates action
live performances go into this bucket as well
Job titles that stay stubbornly human:
Nurse / physical therapist / dental hygienist
Electrician / HVAC tech / plumber / trades
Teacher (especially early childhood)
Firefighter / EMT / Police
Therapist / counselor
And then there’s the category nobody talks about enough:
More valuable because of AI
Work that helps humans make sense of chaos and act anyway.
translation (not languages per say but rather help translate skills from one career to another)
trust-building in saturated, AI-generated markets
domain-specific implementation (actual workflows, n8n…or if you’ve heard me speak Nancy 8 Nancy)
“picks and shovels” for adoption: training, governance, auditing, enablement
high-integrity relationship work where you MUST have a human in the loop.
That’s the repricing direction: toward the irreducibly human.
Not “artists will save us…..and not “robots will take everything.”
More like: if a machine can do it for pennies, your value won’t be “doing it.”
Your value will be directing it, verifying it, integrating it, and doing the parts that require presence, trust, and judgment.
Which brings us to the line I can’t stop thinking about:
You’re not getting replaced. You’re getting repriced.
The market is recalculating what your time is worth, based less on your title and more on three things:
How fast you learn
How well you integrate tools into real workflows
Whether your value is replicable for pennies per hour
That’s not a layoff notice…yet.
It’s a negotiation that already started and what’s crazy is that most people don’t even realize they’re at the table.
Here’s what I mean, I’m seeing this every week.
I spend a lot of time teaching people and executive teams to use AI, and the split is always the same:
About 10% lean in.
They find two or three workflows, run reps, and within weeks they’re operating at a level that used to take years to reach.
The other 90% nod politely, try the tool once, get something that reads like a well-meaning alien who studied human communication through airport bookstore bestsellers… and quietly go back to their old system.
Both groups still have jobs today.
But one group just got more valuable.
The other group got easier to replace.
That’s repricing, and it’s already happening.
Stop consuming fear and convert it into a plan.
Doom essays will keep coming and they’ll be well-written, logically tight and they’ll stack dominoes in devastating sequence.
And they will almost never account for the only variable that actually matters:
what YOU do next.
Most people consume fear and go to bed anxious.
A smaller group opens a LLM and turns the scenario into a strategy session.
That gap, fear consumption vs fear conversion, is the repricing.
So here’s the challenge:
Next time a headline makes your stomach drop, don’t share and spiral…
Open your AI of choice and ask the grown-up question:
“If this is directionally right, where does value move, and where do I stand?”
You might not love every answer.
But you’ll be at the table instead of on the menu.
And right now, that’s exactly where you want to be.
This is such a powerful reframe. “You’re not getting replaced to you’re getting repriced” is the conversation most people don’t realize they’re already in.