College in the Age of AI: Is a Degree Worth It?

College Still Pays More, But the Gap Has Stopped Growing

Start with the basics. College grads earn more money — a lot more.

According to the BLS, workers with a bachelor’s degree earn a median of $1,543 per week, compared to $966 for workers with only a high school diploma. That’s a 60% difference. It’s real and it hasn’t gone away.

But here’s what has changed: the gap has stopped growing. Harvard economists Lawrence Katz and Claudia Goldin found that the pay advantage for college grads has barely moved in 25 years, and the SF Fed linked that plateau to employers needing fewer degree-holders for roles that AI can now handle.

What this means for families thinking about college ROI: the diploma isn’t the differentiator it used to be. It still opens doors. It just doesn’t automatically determine what’s on the other side of them. That depends much more on what you studied and where you’re trying to go.

A degree still pays. The question is which degree, aimed at which job.

New Grads Are Struggling More Than They Used To

This is the part most families don’t see coming.

The job market for new college grads has gotten harder. According to the New York Fed, the unemployment rate for graduates ages 22 to 27 hit 5.7% in early 2026, up from a long-term average of 4.5%. That might sound small, but it represents a meaningful shift in how quickly new grads find work — and what kind of work they find.

Two forces are driving this. The first is AI. A Dallas Fed study found that in fields like finance, insurance, and professional services, AI is cutting the number of entry-level roles while pushing up pay for experienced workers. Junior positions are disappearing. Senior positions are paying more. New grads are landing in a market where the bottom rungs of the career ladder are being removed.

The second force is less obvious: remote work. A New York Fed study found that the rise of distributed teams has made it harder for managers to train and mentor new hires, and companies have responded by becoming more reluctant to hire entry-level workers at all. The study estimated that remote work accounts for about 64% of the recent rise in youth unemployment — more than AI.

Put those two forces together and you get the situation new grads face today: fewer entry-level positions, harder to get mentored, slower to advance. None of this is permanent — but it makes the choice of major and career path more consequential than it was even five years ago.

Knowing How to Use AI Now Pays More Than a Degree Alone

Here’s the shift that most families haven’t fully absorbed yet.

A World Economic Forum analysis of more than 10 million job postings found that workers with AI skills earn a 23% wage premium over comparable workers without them. A bachelor’s degree alone, by comparison, adds only about 8% to weekly pay.

PwC’s Barometer, which studied more than one billion job postings across six continents, found an even wider gap — workers with AI skills earning 62% more than peers in the same roles who lack those skills.

Think about what that means in practice. Two graduates, same major, same school. One has built real competency with AI tools relevant to their field. The other hasn’t. By the numbers, they’re entering the job market at very different starting points.

The degree isn’t worthless — it’s still the baseline. But the skill layer on top of the degree is what’s now driving the biggest pay differences. That makes choosing a program that actively builds those capabilities a much more important decision than it used to be.

Some Majors Are at Risk. Others Aren’t.

Not every major faces the same pressure. The BLS 2024–34 Employment Projections put specific numbers on where the risk lands — and where the opportunity is.

Majors under pressure tend to lead to jobs built around tasks AI handles well: sorting information, producing templated documents, basic financial modeling, routine data entry. According to BLS projections, office and administrative support roles will decline 3.9% over the next decade, with AI and automated systems cited as the direct cause. Paralegals face flat hiring as AI takes over contract review, discovery, and research work. Customer service reps, billing clerks, and procurement clerks face similar headwinds. Students entering general business, basic finance, marketing, and journalism are stepping into shrinking entry-level markets.

Majors holding their value share a common thread: the work requires a human being physically present, making real-time judgment calls, or carrying personal accountability for the outcome. AI can assist those workers, but it can’t replace them.

Healthcare is the starkest example. The BLS projects nurse practitioners will grow 40% by 2034, adding 128,400 jobs. But that growth projection actually understates the opportunity, because the U.S. is already running a shortage of roughly 295,800 nurses right now, with nearly 1 in 10 RN positions sitting vacant at the average hospital. Mercer estimates 6.5 million workers will exit healthcare support roles by 2026 with only 1.9 million replacements ready to step in. This is not a field where new graduates compete for scarce positions. It’s a field that is actively running out of people.

Skilled trades tell a similar story. According to ABC’s analysis, the construction industry needs 349,000 net new workers in 2026, climbing to 456,000 by 2027. JLL estimates 2.1 million positions — electricians, HVAC technicians, plumbers, pipefitters — could go unfilled by 2030. Ford alone reports 5,000 open roles it can’t fill, at salaries reaching $120,000. The reason wages in the trades are rising isn’t complicated: there aren’t enough qualified workers and no amount of AI fixes that.

Data science and cybersecurity occupy a different position. These are fields where AI is actually generating demand, not reducing it. Data scientists are projected to grow 33.5% by 2034, adding 82,500 jobs. Security analysts are projected to grow 28.5%, adding 52,100 jobs. Every organization deploying AI needs people who understand how to build, manage, and protect those systems — and that population is in short supply.

The old “STEM is safe, humanities are not” framing is too blunt to be useful. The real dividing line is simpler: jobs that require physical presence, human judgment, or personal accountability are growing. Jobs built around producing written or digital output are shrinking.

How to Pick the Right Degree

The name on the diploma matters less than the job it leads to, and whether that job is one AI replaces early or late.

Most students and parents approach this backward. They pick a school or a major first, then figure out where it leads. The more reliable approach is to identify the career destination first — specifically, the actual first job after graduation — and work backward from there.

Here’s a simple four-step check that applies to any degree:

Step 1: Find the actual first job. Not the five-year goal. Not the aspirational title. The specific role a graduate steps into the year after finishing. What does that person do on a Tuesday afternoon?

Step 2: Check how much of that work AI can do. Sorting, summarizing, drafting, routing, and processing? High risk. Direct patient care, physical installation, legal strategy, complex negotiation, novel problem-solving? Lower risk. Be honest here — most people overestimate how unique their prospective job is.

Step 3: Look at what skills the program actually builds. There’s a difference between a degree that teaches a subject and a degree that builds capabilities. Does the program develop data fluency, communication, domain expertise, or technical skills that make AI tools more powerful in trained hands? Those graduates don’t compete with AI — they outperform workers who don’t know how to use it.

Step 4: Check where graduates actually land. Schools have an incentive to make their outcomes look good. Look at third-party data. Where do graduates work one year out? What are they earning? Does the school have documented employer relationships in the specific field your student is targeting?

This check works for any major, any career goal, any student. The answers will look different for every family — but the questions are always the same.

So Is a Degree Worth It?

Yes. But the honest answer has gotten more complicated.

The earnings gap between degree-holders and non-degree-holders is real, documented, and still significant. But that gap is no longer guaranteed by the credential alone. It depends on the field, the program, the skills developed, and whether the career path leads to work AI is augmenting or work AI is replacing.

Students who come out ahead over the next decade will be the ones who made that evaluation deliberately — starting with where they want to go and building a plan backward from there. Not the ones who picked a name they recognized and hoped for the best.

That’s exactly what Waystone is built to help with. Enter your student’s academic profile, activities, and career direction. Waystone builds a full planning report: school list, activity gaps, tuition pathways, and course priorities, all designed around where your student is actually trying to go. Start here — it takes about ten minutes.

Key Takeaways

• College grads earn about 60% more per week than high school grads, but the gap stagnated around 2000
• Graduate unemployment hit 5.7% in early 2026, up from a long-term average of 4.5%
• Remote work accounts for about 64% of the rise in youth unemployment — not AI
• AI skills carry a 23% wage premium vs. 8% for a bachelor’s degree alone
• Office support roles are projected to decline 3.9% by 2034, with AI cited as the direct cause
• Nurse practitioners are projected to grow 40% by 2034, into a field short 295,800 workers
• Construction needs 349,000 new workers in 2026 alone; 2.1 million trades jobs could go unfilled by 2030
• Data scientists are projected to grow 33.5% and security analysts 28.5% by 2034
• Pick the career target first, then work backward to the degree and school