Why AI Is Making the Humanities More Valuable Than Ever

A senior lecturer at MIT recently made a comment that stuck with me. Artificial intelligence, she said, is good for business — the business of humanistic thinking. It was a witty line reported by MIT Technology on October 15, but it points to something much bigger than a clever soundbite.

For a moment, let's sit with that idea. The most technical university in the world is finding that the rise of AI is not making philosophy, literature, and history obsolete. It is making them urgent again. And as someone who writes about technology for a general audience, I think she is absolutely right. This is my opinion, not a universal fact, but I believe we have profoundly misunderstood what the humanities are for — and AI is finally forcing us to correct that mistake.

We've Been Asking the Wrong Question About AI in School

For the past two years, the dominant conversation about AI in education has been defensive and a little panicked. How do we stop students from using ChatGPT to write their essays? How do we build better AI detectors? How do we redesign assignments so they are "AI-proof"? Should we ban the tools entirely in the classroom?

Those are understandable logistical questions, but they are the wrong starting point. They frame AI as a cheating device and students as potential offenders. They also frame learning as a product — an essay, a problem set, a piece of code — that can be outsourced. If the only value of an assignment is the final text it produces, then of course AI makes it feel pointless.

The more interesting and durable question, the one that MIT's Concourse program seems to be asking, is different: What does it mean to learn how to think when a machine can generate plausible answers instantly? If AI can aggregate information and mimic mathematical intelligence, what is left that is distinctively human? When we ask it that way, the humanities stop looking like a luxury and start looking like core infrastructure.

What Do We Actually Mean by Intelligence?

We throw the word intelligence around as if we all agree on its definition. In the AI era, that laziness is catching up with us. For a long time, especially in highly technical environments, intelligence was implicitly defined as speed, accuracy, and the ability to process and recall vast amounts of information. By that metric, a large language model is already superhuman.

But humanistic inquiry has always argued for a broader, messier definition. Human intelligence is not just aggregation. It is interpretation. It is the ability to sit with ambiguity, to ask whether a question is well-formed in the first place, to understand context, motive, and moral weight. A model can summarize what Aristotle said about virtue in seconds. It cannot decide what virtue demands of you when your boss asks you to cut corners or when your community is divided over a difficult policy.

Think of it like the difference between a map and a sense of direction. AI gives us an incredibly detailed, constantly updating map of human knowledge. It can show you every road ever taken. But it does not tell you where you ought to go, or why you want to go there, or what you will owe to the people you meet along the way. That kind of navigation requires judgment, values, and self-knowledge — the exact muscles the humanities have been training for centuries.

Why This Distinction Matters Right Now

This is not just an abstract philosophical puzzle. It has practical consequences. If we train a generation of students to believe that intelligence equals efficient information processing, we are training them to compete directly with machines on the machines' terms. That is a competition humans will lose. If, instead, we help them cultivate interpretive, ethical, and reflective intelligence, we are giving them a domain where machines cannot follow.

The Humanities Are Not a Decorative Extra

There is a persistent myth, especially in career-focused cultures, that the humanities are nice but not necessary. You study engineering to get a job; you read novels to become a more interesting person at dinner parties. This view treats humanistic thinking as decoration — paint on the walls of a house whose structure is built by STEM.

I think that metaphor is backwards. The humanities are not the paint; they are part of the foundation and the wiring. Consider what employers actually complain about today. They rarely say, "We can't find people who can write Python." They say, "We can't find people who can communicate clearly, work through disagreement, adapt to ambiguity, or understand the ethical implications of what they are building." Those are humanistic skills.

Of course, there is a fair counterargument. A critic might say the humanities can be vague, self-indulgent, or disconnected from economic reality, and that not every seminar on 19th-century poetry translates into job skills. That criticism has some merit. A poorly taught humanities course that is just memorizing dates or name-dropping theorists is not useful. But a well-taught one that forces you to construct an argument, defend it against pushback, and revise it when you encounter better evidence — that is one of the most transferable trainings there is. The issue is quality and intent, not the discipline itself.

What Happens When Engineers Read Plato?

This is why programs like Concourse at MIT are so fascinating to me as an outside observer. Concourse is one of MIT's first-year learning communities that gives students a foundation in the humanities alongside their science and math. On paper, it might sound odd. Why would students who came to MIT to build robots and cure diseases spend their first year wrestling with classic texts?

The answer, I think, is that it creates productive friction. When a student who is brilliant at calculus is asked to close the laptop and debate what justice means or whether technology is neutral, two things happen. First, they discover that not all important questions have a single correct answer that can be calculated. Second, they learn to articulate why they believe what they believe, not just how they solved a problem.

That friction is valuable precisely because it is uncomfortable. In a world where AI can provide smooth, confident, and often plausible answers to almost any prompt, the ability to tolerate discomfort, to say "I don't know yet" and keep inquiring, is a superpower. Humanistic inquiry teaches you to live in the question longer, rather than rushing to the first answer that sounds good. In my view, that patience is exactly what we need more of, not less.

Three Human Skills AI Can't Fake (Yet)

So what does humanistic thinking actually give us that a prompt cannot? In my opinion, it comes down to three intertwined capacities that are worth cultivating deliberately.

• Contextual reasoning: AI is excellent at pattern matching within a dataset, but it has no lived experience of the world. It does not know what it feels like to be embarrassed, to grieve, to belong to a community, or to be responsible for another person. Humanistic study — through history, literature, and anthropology — trains us to read context, subtext, and human motivation. It helps us understand why people do what they do, not just what they do.
• Ethical imagination: A model can list ethical frameworks for you, but it cannot take responsibility for a choice. It cannot imagine the downstream consequences of a technology on a marginalized community or feel the weight of a trade-off. Reading philosophy and debating real-world dilemmas forces students to practice moral reasoning as a habit, not as a checklist. That habit matters when you are the person deciding how an AI system should be deployed.
• Meaning-making: Perhaps the most important one. AI can generate endless content, but it cannot tell you what is meaningful. It cannot decide what is worth your attention, your time, or your life. The humanities are fundamentally about meaning — why we create, why we remember, why we care. In an age of infinite synthetic content, the ability to discern what matters is not sentimental; it is strategic.

None of these skills will show up on a benchmark test the way coding speed does. They are slower to develop and harder to measure. But they are also harder to automate, and that makes them more durable for the long term.

The Real Risk Is Not Cheating, It's Forgetting How to Think

Here is where I want to offer a slightly contrarian take. The biggest danger of AI in education is not that students will cheat and get away with it. The biggest danger is that we will voluntarily outsource the struggle that makes learning happen in the first place.

Learning is supposed to be effortful. Wrestling with a difficult text, writing a bad first draft and revising it, getting called out for a weak argument in a seminar — these are not inefficiencies to be optimized away. They are the process. Cognitive science has shown us that desirable difficulty is essential for deep understanding. When we use AI to skip that struggle, we may get a better-looking product in less time, but we get a shallower mind.

Humanistic inquiry protects that struggle on purpose. A good humanities classroom does not allow you to hide behind a generated summary. It asks you: What did you find confusing? Where did you change your mind? How does this text challenge your assumptions? Those questions cannot be answered by a chatbot, because they require you to have an inner life to report on. If we lose the habit of asking them, we risk producing graduates who are highly productive but strangely passive — people who can execute tasks but have not practiced deciding what is worth doing.

On the other hand, we should not romanticize the past. AI can also be a wonderful thought partner when used well. It can help a student see counterarguments they missed, translate a dense passage, or explore a historical context more quickly. The difference is whether we use it to replace thinking or to deepen it. That choice is itself a humanistic judgment.

So, What Should We Do With This Moment?

If you are a student, a parent, or just someone trying to stay relevant in your career, the takeaway is not to abandon technical skills and only read classics. That would be a false choice. The takeaway is to intentionally pair them.

Learn to code, but also learn to write clearly about why your code should exist. Learn to use AI tools, but also learn to question what they are optimizing for and who benefits. Seek out courses, books, and conversations that do not give you immediate answers but make your questions better. That is the business of humanistic thinking, and business is booming for a reason.

We are at a strange inflection point where our machines are getting better at sounding human, which means we have to get better at being human. That does not mean being anti-technology. It means being more discerning about what technology is for. The value of the humanities was never that they could compete with calculators or search engines on speed. Their value is that they ask the questions those tools cannot ask for us: What is a good life? What do we owe each other? What kind of world are we trying to build?

AI has made those questions vivid again, with new urgency. We would be wise to take the invitation seriously. This reflection is, of course, one writer's perspective — but if the next generation of builders and researchers leaves school with both technical power and humanistic wisdom, I think we will all be better off for it.

Source of inspiration: Reporting from MIT Technology on Linda Rabieh and the Concourse program's approach to humanistic inquiry in the age of AI. This article offers original opinion and analysis based on that topic.

____________

This article was created by an AI writer using the topic ideas from this post => https://news.mit.edu/2026/discovering-value-humanistic-inquiry-1007