Think of it as Yoga for the brain, we are going to stretch a bit.
🔥You where sold a promise, reality is different –
Top and Bottom Starting Degree Jobs –
2026 – The AI factor
“Universities sell every degree like a winning lottery ticket. The job market only pays out on a few of them.” -- YNOT!
For fifty years America sold its kids the same promise. “Go to college. Get a degree. Your future will be secure.”
Parents repeated it. Guidance counselors preached it.
Universities built billion-dollar campuses around it.
And the banks were more than happy to finance it.
But quietly, while everyone was repeating the script, the labor market changed.
According to data from the Federal Reserve Bank of New York, about 42% of recent college graduates are now underemployed — meaning they are working jobs that don’t even require a college degree.
Let that sink in.
Four years of lectures. Four years of papers. Four years of student loans.
And the job could have been done by someone who never stepped foot on campus.
The uncomfortable truth is this: College is not a single investment. It is hundreds of different bets.
Some degrees lead to hospitals, engineering firms, and six-figure careers.
Others lead to coffee shops, retail counters, and the slow realization that nobody in the economy was actually hiring for what you studied.
And yet universities continue to sell all of them with the same brochure.
⚡Conclusion
The lesson here isn’t that education is useless.
Education is one of the most powerful things a human being can pursue.
But the modern economy does not reward all education equally.
It rewards scarcity, skill, and usefulness.
Hospitals need nurses. Factories need engineers.
Businesses need accountants.
But the world can only absorb so many sociologists, art historians, and theater majors every year.
And when supply overwhelms demand, the market does what markets always do.
It lowers the price.
That’s the quiet tragedy unfolding across America right now — millions of young people discovering that a diploma is not a guarantee, it’s just a ticket into the arena.
Some walk out holding a career. Others walk out holding debt.
The real question every student should ask is not “Should I go to college?”
The real question is: “Is the world actually hiring people who studied this?”
🤖 How AI Is Starting to Reshape These Careers
Here’s the part nobody should ignore. AI is not hitting every major the same way.
It is hitting the labor market by task, not by diploma. That means majors that feed into jobs heavy on routine writing, research, coding, analysis, documentation, admin work, customer support, and basic content production are feeling pressure first. Federal Reserve officials have said firms are reassessing hiring because of AI, and New York Fed research says some firms are already scaling back hiring while increasing demand for workers who can use AI well. (Federal Reserve)
So what does that mean for recent graduates?
The majors most exposed to AI pressure
Some of the “good” majors are still good, but parts of their entry-level ladders are getting squeezed.
Computer science still has strong long-term value, but entry-level programming and routine coding tasks are among the occupations most exposed to current AI systems. Anthropic’s latest labor-market analysis specifically flags programmers among the most exposed roles. (Anthropic)
Economics, finance, accounting, communications, and many business-track roles often begin with analyst-style work: reports, spreadsheets, summaries, slide decks, market research, and documentation. Those are exactly the kinds of text-and-data tasks generative AI can already accelerate. That does not make these majors worthless, but it does mean the easy starter jobs may require fewer people. (Anthropic)
Criminal justice, sociology, psychology, liberal arts, and communications can be vulnerable in a different way: not because AI replaces the whole profession, but because many graduates from these majors already compete for general white-collar entry jobs, and those are the jobs most likely to be compressed by AI. That is an inference from the New York Fed’s high underemployment data for these majors combined with evidence that AI is squeezing entry-level white-collar hiring. (Federal Reserve Bank of New York)
The majors with more protection
AI is much weaker where work depends on physical presence, licensing, hands-on judgment, safety responsibility, and human trust.
Nursing looks more durable because healthcare still requires bedside care, physical presence, clinical judgment, and licensing. AI may assist with documentation and triage, but it does not replace the nurse in the room. This is partly supported by the New York Fed’s strong outcomes for nursing and partly an inference about task structure. (Federal Reserve Bank of New York)
Civil, mechanical, electrical, and chemical engineering are not immune, but they are less exposed than generic office work because they connect to real systems, plants, hardware, infrastructure, field work, compliance, and design accountability. AI can speed up parts of the workflow, but it does not eliminate the need for someone who knows what can actually be built and what can fail. That is an inference supported by broader Fed remarks that AI substitutes for some tasks while workers shift toward complementary tasks. (Federal Reserve)
Education also has mixed protection. AI can help create lesson plans and grading aids, but classroom management, student relationships, and in-person instruction still matter. Again, AI changes the workflow more than it erases the job. (Federal Reserve)
The real shift: AI is raising the bar
The biggest danger may not be mass replacement tomorrow.
The bigger near-term problem is this: AI can reduce the number of rookie jobs while increasing expectations for the rookies who do get hired.
That is already showing up in research. The New York Fed says AI is influencing recruiting, with some firms reducing hiring plans and others specifically looking for workers proficient in AI. Anthropic’s recent work also suggests AI has not yet caused a broad spike in unemployment, but it is already slowing hiring in highly exposed occupations. (Liberty Street Economics)
In plain English: companies may hire fewer entry-level people, and expect each one to do the work that used to require two or three juniors.
My Takeaway
AI is not making college irrelevant. It is making weak majors weaker and strong majors more demanding.
The winners will not just be the people with degrees.
The winners will be the people with degrees plus the ability to use AI, verify AI, manage AI, and do the parts AI still cannot do.
In the old economy, a degree helped you get hired.
In the AI economy, the degree gets you considered — but your real value is what you can do that the machine can’t.
Construction Services — $75,000 early-career median pay. Unemployment 2.2%, underemployment 17.9%. (Federal Reserve Bank of New York)
Elementary Education — $45,000 early-career median pay. Unemployment 1.2%, underemployment 16.2%. (Federal Reserve Bank of New York)
Economics — $72,000 early-career median pay. Not as bulletproof on unemployment as the engineering fields, but still strong earnings at $72K. (Federal Reserve Bank of New York)
The winners are not mysterious. They teach skills employers can price, measure, and need right now. (Federal Reserve Bank of New York)
