How Duke Supplemental Essay Asks Students About AI Decision Making
Duke University throws a curveball at applicants that most students don’t see coming. Buried in their supplemental essays sits a question about artificial intelligence that forces you to think critically about decisions, ethics, and your own values.
This isn’t your typical “why Duke” prompt. It’s harder - and honestly? That’s what makes it interesting.
What Duke Actually Asks About AI
Duke’s supplemental essay question about AI decision-making typically asks students to consider scenarios where artificial intelligence makes choices that affect people’s lives. The prompt wants you to wrestle with real ethical dilemmas-not regurgitate talking points about robots taking over the world.
The question might look something like this: “Describe a situation where you think AI should or should not be trusted to decide. Explain your reasoning.
250-300 words - that’s it. You need to pack a punch.
Here’s what Duke’s admissions committee is actually evaluating:
- Your ability to think through complex problems
- How you weigh competing values
- Whether you can articulate nuanced positions
They’re not looking for a tech genius. They want a thoughtful person who engages with difficult questions.
Step 1: Choose Your AI Scenario Carefully
Don’t pick the obvious examples. Admissions officers have read hundreds of essays about self-driving cars and facial recognition. Those topics aren’t bad-they’re just overused.
Instead, consider these angles:
- AI in college admissions (meta, yes, but genuinely interesting)
- Algorithms that determine bail amounts in criminal cases
- AI-powered content moderation on social platforms
- Machine learning systems that diagnose medical conditions
- Automated hiring tools that screen resumes
Pick something you actually care about. Your genuine interest will show through the writing.
Why this matters: Admissions readers can spot a student writing about something they researched for 20 minutes versus something they’ve genuinely pondered. Authentic curiosity translates to engaging prose.
Step 2: Take a Clear Position (Then Complicate It)
Here’s where most students mess up. They try to play it safe by saying “AI has both benefits and drawbacks. " That’s not an essay - that’s a throat-clearing exercise.
Take a stance - be specific.
Bad approach: “I think AI in healthcare has pros and cons that we need to consider carefully.”
Better approach: “AI should diagnose skin cancer from images, but a human oncologist should always deliver that diagnosis to the patient.”
See the difference? The second version draws a clear line. It shows you’ve thought about where human judgment remains essential.
But don’t stop there. After stating your position, acknowledge the strongest counterargument. This shows intellectual honesty.
You might write: “Some argue that AI delivers diagnoses more consistently, without the emotional hesitation that sometimes delays human doctors. They have a point. But medicine is more than about accuracy-it’s about care.
That kind of nuance impresses admissions committees.
Step 3: Connect It to Your Own Experience
Duke isn’t asking for a policy paper. They want to know how you think about these issues.
Maybe you’ve used AI writing assistants for school and noticed they sometimes suggest factually incorrect information. Maybe a family member was denied a loan by an automated system. Maybe you’ve watched your younger sibling interact with AI chatbots in ways that concern you.
Personal stakes make essays memorable.
One approach that works well: Start with the abstract principle, then zoom into a specific moment that shaped your thinking.
“Last March, my grandmother received a letter denying her insurance claim. The denial was generated automatically. No human had reviewed her case. She spent three weeks on hold, trying to reach someone who could actually look at her situation. That experience changed how I think about algorithmic efficiency.
Now you’ve grounded your essay in something real.
Step 4: Demonstrate You’ve Done Your Homework
Mention specific details that show you understand the technology. You don’t need a computer science degree-just do enough research to speak credibly.
Instead of writing: “AI makes mistakes sometimes.”
Write: “Training data bias led Amazon’s hiring algorithm to downgrade resumes that included the word ‘women’s’-like ‘women’s chess club captain. ’ The system learned from 10 years of hiring decisions that favored men.
That specific example demonstrates genuine engagement with the topic.
Troubleshooting tip: If you can’t find specific examples, you haven’t researched enough. Spend 30-45 minutes reading articles from Wired, MIT Technology Review, or academic publications. The specifics will come.
Step 5: Avoid These Common Mistakes
**Don’t be preachy. ** Admissions officers aren’t looking for someone who lectures them about the dangers of technology. They want curious thinkers, not moral crusaders.
**Don’t pretend expertise you don’t have. ** Phrases like “as someone who deeply understands machine learning” will backfire unless you’re genuinely building AI systems.
**Don’t ignore tradeoffs. ** Every position on AI involves tradeoffs. Acknowledging them shows maturity.
**Don’t write what you think they want to hear. ** Duke accepts students with diverse viewpoints. Trying to guess the “right” answer produces bland essays.
What a Strong Response Actually Looks Like
Here’s a framework for structuring your 250-300 words:
- Opening hook (1-2 sentences): A specific observation or question that grabs attention
- Your position (2-3 sentences): Clear stance on when AI should or shouldn’t decide
- Supporting reasoning (3-4 sentences): Why you hold this position, with specifics
- Counterpoint acknowledgment (1-2 sentences): The strongest argument against your view
- Personal connection (2-3 sentences): How this relates to your experience or values
- Closing insight (1-2 sentences): What this reveals about your thinking
That’s approximately 250 words if you’re concise.
Final Thoughts on the Duke AI Essay
This prompt separates students who engage with ideas from those who just complete assignments. Duke wants undergraduates who will contribute to classroom discussions, challenge assumptions, and think carefully about technology’s role in human life.
The question isn’t really about AI. It’s about how you reason through uncertainty. How you balance efficiency against humanity. How you hold strong opinions loosely enough to consider other perspectives.
Start drafting early. Your first version will probably be too general or too preachy. That’s normal. Revise until you’ve said something specific, personal, and genuinely interesting.
And remember-you’re not trying to solve AI ethics in 250 words. You’re trying to show how you think. That’s something only you can do.