Back to the blog

What Should They Study Now?

What should students study in the age of AI? The answer is shifting. While STEM remains important, skills like critical thinking, communication, and adaptability are becoming more valuable. This piece explores why liberal arts and a broad, flexible skill set may offer the…

Erik CharlesMarch 17, 20264 min read
What Should They Study Now?

Rethinking College Majors in the Age of AI

By Erik Charles, guest blogger

The Ways We Work

Volume 2, Number 4

The Question Every Parent Is Asking

College acceptances are rolling in. Along with them come familiar questions: cost, location, what happens to the empty bedroom, and most importantly:

What should they study?

For years, the answer felt straightforward. STEM was the default path. It was practical, durable, and safe.

But that consensus is starting to shift.

A new perspective is emerging from the very people building the AI-powered future. It challenges one of the most deeply held assumptions about education and work.

Rethinking the “Safe” Path

I studied Political Science and Economics and began my career in management consulting before moving into software startups. My older son studied political science at Berkeley and now works in technology sales. My younger son studied international relations at Stanford and works in operations in commercial real estate.

None of these paths was traditionally “technical.” All of them are working.

That is not an exception anymore. It is becoming a pattern.

Leaders at the forefront of AI are increasingly vocal. The skills that will matter most are not purely technical. They are human.

Daniela Amodei, President and co-founder of Anthropic, recently said that studying the humanities will be “more important than ever.” Her reasoning is simple. AI excels at structured, technical tasks. What remains uniquely valuable is understanding people, how we think, decide, communicate, and relate.

The Myth of the “Dead-End” Degree

The idea that the humanities lead to dead ends has been quietly disproven.

Some of the most influential builders and investors in the modern economy came from liberal arts backgrounds:

  • Reid Hoffman (LinkedIn), Philosophy

  • Stewart Butterfield (Slack, Flickr), Philosophy

  • Carl Icahn, Philosophy

  • George Soros, Philosophy

  • Mary Shea (Meerkat), Musicology & Ethnomusicology

These are not edge cases. They are signals.

Even technical leaders are reinforcing this shift. Goldman Sachs’ CIO recently advised his daughter to study philosophy, even if she wants to pursue engineering, because it develops the ability to build clear mental models and think precisely.

That combination is becoming more valuable, not less.

Where Liberal Arts Meets AI

The future is not a choice between technical and non-technical paths. It is about the ability to operate across both.

This is where liberal arts graduates increasingly stand out.

Prompting is applied in the liberal arts. The ability to ask the right question, understand nuance, and guide an AI system toward useful output is fundamentally a language and reasoning problem.

Judgment becomes the differentiator. As AI generates more content, the bottleneck shifts from creation to evaluation. What is correct? What is biased? What is actually useful?

Translation is the new power skill. Organizations need people who can bridge technical systems and human needs. They translate complexity into clarity.

These are not new skills. They are foundational ones.

They closely mirror the attributes described in Future-Proofing Leadership in the Age of AI. Curiosity, critical thinking, and conviction are becoming defining advantages in an AI-driven world.

The Return of the Generalist

The phrase “Jack of all trades” has evolved over centuries, from compliment to criticism and back again:

“…but oftentimes better than a master of one.”

That idea feels especially relevant now.

As AI automates narrow expertise, depth in a single domain becomes more fragile. Breadth, paired with the ability to adapt, makes for greater durability.

The goal is not to be mediocre at everything. It is to be capable across enough domains to pivot when the landscape shifts.

Because it will.

Building a More Resilient Skill Set

If the goal is adaptability, then education should expand beyond a single track.

A few examples:

  • Take an improv or communication class to build real-time thinking and adaptability

  • Study art or design to develop taste and discernment in an AI-saturated world

  • Learn programming fundamentals to understand how systems work

  • Build financial literacy to connect your work to business outcomes

  • Invest in networking and relationship-building

  • Take creative writing to strengthen clarity and originality

These are not “extra” skills. They are compounding ones.

They align directly with the qualities that define future-ready leaders. These are people who can think independently, direct AI effectively, and operate with judgment in ambiguous environments.

Final Frame: The Value of the “Random” Path

So what should students study?

A better question might be:

How do they build a path that keeps working as the world changes?

In a moment where yesterday’s hard skills are being encoded into software, the most durable advantage lies in perspective, adaptability, and the ability to connect ideas across domains.

Encourage exploration. Encourage breadth. Encourage curiosity.

The goal is not just to prepare for a first job. It is to build a career that evolves.

Because in the end, the advantage does not belong to the person who masters a single trade.

It belongs to whoever knows which trades are worth mastering next.

👉 From Mary: At Meerkat, we’re sharing forward-looking perspectives through The Ways We Work series.

➡️ Follow our journey at trymeerkat.ai and connect with us on LinkedIn to stay up to date on new insights, stories, and resources.

➡️ Download Meerkat here. Use my referral code: 680016.