Does a Philosophy degree still mean “would you like that with fries?”

Photo by David Foodphototasty on Unsplash

For a long time,Philosophy (and other arts and social sciences) as a discipline had branding issues.

Tell someone you studied Philosophy or any of the Social Sciences and there was a decent chance they would respond with the classic joke: “So… would you like that with fries?”

It was admittedly an unkind joke, yet also a very efficient one. In one sentence, it managed to compress an entire worldview: that education is mainly about employability, employability is mainly about technical skill, and technical skill is mainly about doing something obviously useful enough that your relatives can explain it during Chinese New Year.

Then AI arrived, and suddenly the joke became less funny.

Because in an AI-shaped economy, the person who knows how to ask better questions may become more valuable than the person who only knows how to produce better answers.

The Strange Return of Asking Good Questions

For decades, we trained students and workers to become good answer machines.

Memorise the formula ⇒ apply the framework ⇒ follow the process ⇒ pass the exam ⇒ clock-in on time ⇒ clear your tickets ⇒ submit the deck ⇒ smile politely when someone says, “Can you just make this one last small change?” (spoiler: it never was the last)

This worked reasonably well in a world where information was scarce, expertise was locked inside institutions, and workers were rewarded for knowing the right procedure.

And them came AI which changes the shape of work: because it makes answers cheap.

They’re neither perfect answers nor reliable answers, definitely not answers you should blindly copy and paste unless you enjoy potential professional self-sabotage. But yet we cannot deny, answers are now much easier to generate.

• A student can ask an AI to explain photosynthesis
• A marketer can ask for ten campaign angles
• A SWE can ask for a utility function
• A QA can ask for verification of acceptance criteria
• A lead engineer can ask for an architecture revamp
• A manager can ask for a policy template
• HR can ask for a training plan and receive something that looks suspiciously more polished than the average internal document

The bottleneck is no longer just access to information. The bottleneck is judgment.

• What should I ask?
• How should I frame the problem?
• What assumptions are hidden here?
• What would a better version of this answer look like?
• What is missing?
• Who might be harmed if we implement this badly?
• Is this even the right problem?

These are not purely technical questions. These are things that a Philosophy degree prepares you for.

Prompt Engineering Is Basically Philosophy; With 2026 Packaging

“Prompt engineering” sounds very modern, very Silicon Valley, very Anton, very tres-comas.

But underneath the shiny label, much of it is old human craft. It is the craft of asking clearly, of giving context, of knowing when a vague answer is hiding a vague question.

It is the craft of checking whether the machine has confidently produced nonsense with the energy of a man at a kopitiam explaining COE policy after one Tiger or ABC too many.

A good prompt is rarely just a command. It is a structured act of thinking.

You define the role. You give the context. You specify the constraints. You describe the desired output. You ask the model to reason through trade-offs. You test edge cases. You challenge the first answer. You refine.

In other words, you do not merely ask, “Write me something.”

You ask, “Given this audience, this goal, this constraint, this tone, this risk, and this definition of success, what are the strongest options, what are the trade-offs, and what would you recommend?”

That is not button pressing. That is thinking.

And this is where philosophy quietly sneaks back into the room, removes its sandals, and asks why everyone has been pretending it was useless.

Education Has Been Too Obsessed With Producing Correctness

One problem with traditional education is that it often rewards the tidy answer over the interesting question.

This is understandable. It is much easier to grade correctness than curiosity.

A multiple-choice question can be marked quickly. A standard essay can be assessed against a rubric. A math problem has working and an answer. Very efficient. Very scalable. Very Singapore.

But real life is not always so neat.

In the real world, the question is often badly formed. The data is incomplete. The customer does not know what they want. The boss says “AI strategy” but actually means “please make us look modern without changing anything too painful.” The vendor says “seamless integration” and everyone in the room pretends not to hear the faint sound of future suffering.

Young people entering this economy will not just need to know things. They will need to interrogate things.

They will need to question whether an answer is useful, whether it is true, whether it is relevant, whether it is biased, whether it fits the actual context, and whether the problem was framed properly in the first place.

This does not mean facts are irrelevant. Quite the opposite. You cannot think critically from a vacuum. A person with no domain knowledge using AI is like someone confidently ordering fish soup at a prata shop. Possible, perhaps, but something has gone wrong.

The future belongs neither to pure memorisers nor pure vibes-based builders.

It belongs to people who can combine knowledge, curiosity, judgment, and taste.

Training Workers For The AI Economy

For workers, the shift may be even more urgent.

A lot of corporate AI training today still feels like someone discovered ChatGPT last Wednesday and was immediately asked to run a lunch-and-learn.

The training often goes like this: here are ten prompts, here are five use cases, please do not paste confidential data, good luck and may the Gods of productivity look kindly upon you.

That is not enough.

Workers do not just need prompt templates. They need new working habits.

They need to learn how to break messy tasks into smaller parts, how to compare AI-generated options, how to review outputs with domain judgment, and maybe most importantly: when not to use AI.

They need to learn how to document decisions, protect sensitive information, and avoid turning one person’s hallucination into the company’s new operating procedure.

They need to become better questioners.

A finance team using AI should not only ask, “Can you summarise this report for me?” They should ask, “What are the key risks, what assumptions drive the forecast, and what would change the recommendation?”

A customer support team should not only ask, “Can you draft a reply?” They should ask, “What is the customer really upset about, what policy constraints apply, and how do we respond without sounding like a government chatbot having a bad day?”

A leadership team should not only ask, “How can we use AI?” They should ask, “Where does better intelligence actually improve revenue, reduce cost, manage risk, or improve customer experience?”

That last one matters. Otherwise every company ends up with twelve AI pilots, zero business impact, and one poor intern maintaining a spreadsheet called AI_use_cases_final_v7_reallyfinal.xlsx.

Philosophy As A Practical Skill

So, does a Philosophy degree still mean “would you like that with fries?”

Maybe not.

Maybe Philosophy, at its best, is training in precision, ambiguity, ethics, interpretation, and argument. Maybe it is the discipline of noticing that the first question is rarely the best question. Maybe it teaches people to sit with uncertainty long enough to produce a better thought.

That sounds rather useful now.

The AI economy will reward people who can work with machines, but not be overawed by them. People who can ask, challenge, interpret, and decide. People who understand that a fluent answer is not the same as a correct one. People who can move between technical tools and human consequences.

In that world, Philosophy is not a punchline, it is a surprisingly practical foundation.

Of course, nobody is saying every student needs to spend four years debating Metaphysics before they are allowed near a spreadsheet. Singapore already has enough tuition centres; we do not need to have Knowledge & Inquiry for Primary 3 students.

But we may need to gently introduce again the art of questioning into education and training and give it some priority over getting 100/100 in tests.

Teach students how to ask better questions of AI. Teach workers how to challenge generated outputs. Teach teams how to frame problems before rushing to solutions. Teach leaders that “AI adoption” is not a software rollout; it is a thinking upgrade.

Because the future may not belong to the person who can produce the most answers.

It may belong to the person who can frame problems and ask the relevant questions that change realities entirely.

Aijutsu is a Singapore-based fractional technology leadership practice for founders, SME business owners, and lean technology teams that need clarity, change, and delivery confidence across AI, cloud, compliance, infrastructure, operations, and software delivery. Check us out at https://aijutsu.dev