Liberal Arts Majors Are About to Have Their Moment
For two decades, STEM was the safe bet. AI is reversing that—models respond to precise vocabulary, and liberal arts majors have it.
For two decades, STEM has been the safe career bet. Learn to code. Get a technical degree. Liberal arts? Good luck with that.
The joke was always the same: “What do you say to an English major? I’ll have fries with that.”
I think we’re about to see that reverse.
The Hidden Advantage
AI models respond to precise vocabulary. They were trained on expert communities—academic papers, specialized forums, professional discussions. The knowledge is embedded in there, compressed and waiting.
The unlock is knowing the right words.
Who knows the right terminology for visual composition? Art history majors. Who can articulate the contours of an abstract idea? Philosophers and English majors. Who understands the underlying structures of sound and rhythm? Music theorists.
These aren’t soft skills. They’re specialized vocabularies that map directly onto what models can produce.
What This Looks Like in Practice
Anyone can prompt an image generator with “make this look like museum art.”
An art history major prompts differently: “Adjust the lighting to reference Vermeer’s use of diffused northern light. Shift the composition to create more negative space in the upper third—closer to Dutch Golden Age interior scenes.”
The model understands both requests. It produces dramatically different results.
Same with music. An executive might get a passable jingle from Suno. A music producer with background in ethnomusicology prompts: “Blend Gregorian chant modal progressions with 80s synth-pop production. Match the time signatures—both traditions use similar rhythmic structures.”
That producer sees connections invisible to someone without the framework. The model can execute on those connections. It just needs someone who knows they exist.
The Vocabulary Gap
Here’s what I’ve noticed training 200+ professionals on AI:
Technical people often struggle with open-ended creative prompts. They want parameters, specifications, clear requirements. When the task is “make this better” or “make this more compelling,” they freeze up.
Liberal arts people have trained for years in exactly this kind of ambiguity. What makes writing persuasive? What makes an image evocative? What makes an argument compelling? They have frameworks for these questions. More importantly, they have vocabulary.
That vocabulary is the interface to AI capability.
The Depressing Irony
The people best positioned to use AI effectively are often the most resistant to it.
I get it. If you spent four years studying literature or art history, and then watched the job market treat your degree as worthless, you’re not going to feel warmly toward technology that seems to threaten creative work.
But the resistance is backwards. These tools amplify exactly the skills liberal arts education develops: taste, judgment, conceptual thinking, precise articulation.
A philosophy major can prompt an AI to explore the steel-man version of an opposing argument. An English major can guide it toward specific rhetorical effects. An art history major can reference five centuries of visual vocabulary.
These aren’t replaceable skills. They’re the skills that make AI outputs actually good.
If I Were Hiring
I’d prioritize liberal arts backgrounds for AI-intensive roles. Specifically:
- Art history, film studies, visual arts → image generation, video production, design
- English, philosophy, rhetoric → writing, communication, argumentation
- Music theory, ethnomusicology → audio production, sound design
- Anthropology, sociology → research synthesis, cultural context
Then I’d train them on the tools. The technical part is learnable in weeks. The conceptual frameworks take years to develop.
Most companies are doing this backwards. They hire technical people and hope they develop taste. It’s easier to teach prompting to someone with domain expertise than to teach domain expertise to someone who can prompt.
The Real Skill
The value of a liberal arts education was always learning to think in abstractions, make distinctions, see patterns across contexts. The content was secondary.
AI doesn’t replace that. It rewards it.
Models are general-purpose tools waiting for specific direction. The person who can provide that direction—with precision, with vocabulary, with conceptual clarity—gets dramatically better results.
That’s the liberal arts advantage. It’s been undervalued for twenty years. I think that’s about to change.
Interested in how domain expertise translates to AI capability? Book an AI Readiness call - 30 minutes to explore what your team might be missing.