I get it. You’re busy. You have a full-time job, maybe a side hustle, maybe a social life you’re desperately clinging to. And here’s this shiny AI tool that promises to help you write your MBA essays in, like, 20 minutes.

The temptation is REAL.

But I want to talk to you honestly about why leaning on AI as your primary admissions advisor is a genuinely risky move — not because AI is useless, but because the ways it can fail you are subtle enough that you won’t even realize it’s happening until it’s too late.

AI Is Programmed to PLEASE You, Not HELP You

Here’s the thing about large language models that most people don’t fully appreciate: they are optimized to generate responses that feel satisfying. They only stay in business if you use them a lot. They want you to like them. You liking them = their survival.

This means that when you paste your essay draft into ChatGPT and ask “Is this good?”, it is almost certainly going to tell you it’s good. Maybe with a few minor suggestions sprinkled in to seem credible(the exact same suggestions, BTW, that’s it’s giving to everyone else… but more on that later). But it’s not going to look you in the eye and say “Honestly? your career goals are vague and unconvincing and an admissions committee is going to roll their eyes at this.”

A good human advisor — whether that’s an experienced consultant or a brutally honest friend who went to Wharton — will tell you the uncomfortable truth. AI is structurally incentivized NOT to do that.

I’ve spent years reviewing thousands of MBA applications. The feedback that actually helps people is almost never “Great job!” It’s “Your ‘Why MBA’ rationale is weak,” or “This leadership story doesn’t actually show leadership,” or “You sound like you want the degree as a status symbol, not because you have a real plan.” That kind of feedback is hard to give AND hard to hear. But it’s vital that you hear it. And AI will sidestep it every time.

You Don’t Know What You Don’t Know

This one is sneakier, and honestly scarier.

AI makes mistakes. Factual errors, logical gaps, advice that sounds authoritative but is just… wrong. The problem is that if you’re using AI precisely BECAUSE you don’t know much about MBA admissions, you have no filter to catch those mistakes.

Maybe AI tells you that a particular school “Values entrepreneurial backgrounds” when actually that school has been quietly shifting its profile toward finance and consulting candidates for the past three years. Maybe it gives you essay advice that would have been solid in 2020 but misses how a specific school’s prompts have evolved. Maybe it confidently tells you your GMAT score is competitive for a school, but it doesn’t realize that your specific profile is one that needs a score well above the average.

You won’t know. You’ll just… submit. And wonder later why you got dinged everywhere.

This is the “unknown unknowns” problem, and it’s brutal in high-stakes situations like MBA admissions where the margin for error is thin and the consequences of getting it wrong are significant.

(p.s. My favorite recent story of AI leading an unsuspecting applicant astray? In 2025 I received an essay draft for Harvard Business School that cited how excited the person was for the potential to be mentored by famed HBS professor Clayton Christensen. There are 2 problems with this: first of all, I explain in my extensive HBS advice module in ApplicantLab why mentioning a professor may not be an optimal use of space in the HBS essays. The 2nd problem? Well… unfortunately, Clay Christensen passed away in 2020 [RIP to a true legend]. Since the candidate didn’t actually do the research themselves, if they would have relied upon AI they would have submitted a memorable essay indeed – memorable for all the wrong reasons, memorable in the “LOL let’s laugh about this for years to come” way. )

You Risk Sounding Like Everyone Else

Let me paint you a picture of what admissions committees are reading right now.

Thousands of applications. And a growing number of them have a certain… sameness to them. Polished sentences. Logical structure. A kind of frictionless readability that somehow leaves no impression whatsoever. “It’s not robotic; it’s just not quite human” (have you noticed how some AI loves the “It’s not X; it’s Y” construct? I’d guess this turn of phrase has massively increased in MBA essays this year!)

That’s AI-assisted writing. And admissions readers can feel it, even when they can’t definitively prove it. (Actually, in May of 2026 I attended a conference with admissions heads from top 25 programs and many of them pointed out that, just like human writers have a certain subtle “voice” that comes across… AI also has it’s own “voice” that becomes easily identifiable. YOU can’t tell that your essay “Sounds like AI”, but that’s because YOU are NOT reading 5,000+ essays each year!).

The whole POINT of your MBA application is to make a distinct, authentic human impression. Your stories, your voice, your specific weird career detour or the moment you realized you wanted to build something — that stuff is what gets you admitted. When you outsource your voice to a language model trained on millions of generic documents, you get… generic output.

And the generic candidate doesn’t get accepted.

Different schools are looking for genuinely different things. What resonates at Stanford GSB (deeply personal, self-reflective, “What matters most to you and why”) is NOT what works at Booth (analytical, intellectually rigorous, here’s how I think). AI doesn’t really get that nuance. It writes something that sounds like an MBA essay, not something that sounds like YOU applying to THAT specific school.

Ding dong! You Still Have to Do the Interview

Okay, let’s say you somehow navigate all of the above. You use AI heavily, your essays come out fine, you get invited to interview.

Now what?

Because here’s what a lot of people don’t think about: the interview is where everything you supposedly wrote about gets tested in real time. An admissions interviewer is going to ask you to expand on your career goals. They’re going to probe your leadership stories. They’re going to ask follow-up questions that require you to actually KNOW your own application inside and out — and to have genuine conviction about what you wrote.

If AI did the heavy lifting on your essays, you didn’t do the thinking. You didn’t sit with the uncomfortable question of “Wait, do I actually have a clear vision for what I want to do post-MBA?” You didn’t wrestle with which leadership experience actually says something meaningful about who you are. You just… accepted whatever the AI generated and moved on.

And it will show. In the interview, it will absolutely show.

Admissions committees at top schools are increasingly using video components and live interviews precisely because they want to see the real person behind the application. You can’t AI your way through a 30-minute conversation with a Harvard interviewer.

The Work IS the Point

I want to make a slightly bigger argument here, because I think it matters.

The process of deeply reflecting on your career, your experiences, your leadership moments, your goals — that process is VALUABLE BEYOND just getting you into business school.

When you actually sit down and force yourself to articulate why you want an MBA, what you’ve accomplished, what kind of leader you are, what you want your career to look like in ten years — you come out of that process knowing yourself better. You get to school with a plan, instead of flailing around in circles, bouncing from interest to interest (and wasting precious time). You’re sharper in interviews. You’re clearer in conversations with potential mentors or employers (fun fact: MANY ApplicantLab users reach out to us once they’re IN the MBA program to thank us for forcing them to do this work, since they feel that it’s helped them ace their JOB interviews… you know, the whole point of going to the MBA in the first place). You have a story you can actually tell, because you’ve told it to yourself first.

If you skip that work by outsourcing it to AI, you deprive yourself of something genuinely useful. And then you show up to an interview — or, honestly, to business school itself — without having done the thinking that makes the whole thing worthwhile.

I’ve seen this happen. Candidates who had polished applications but couldn’t hold a coherent conversation about their own goals. It’s a real problem.

So What SHOULD You Use AI For?

I’m not saying AI has zero role in this process. That would be silly.

AI can be useful for: – Researching schools (with appropriate skepticism about outdated info) – Checking grammar and basic clarity

What AI should NOT be doing: – Evaluating whether your career goals are actually compelling (as an experiment, I’ve asked it questions about potential career paths for my teenaged son, and sorry but the AI slop it comes back with, while flattering, is also full of b.s.) – Telling you if your stories are the right stories for a specific school (that’s a level of human nuance it pretends to know but doesn’t actually know) – Replacing the strategic thinking about how to position your application – Writing your essays from scratch and calling it done

The difference is using AI as a tool versus using AI as a consultant. One is fine. The other is a gamble with your application.

The Bottom Line

MBA admissions at top programs is genuinely competitive. You’re going up against people who have thought hard about their stories, gotten real feedback from people who know what they’re doing, and done the actual work of self-reflection that makes for a compelling application.You’ll then be competing for jobs against people who enter the program with a clear sense of who they are and where they’re going.

AI will make you feel like you’re doing the work. It will produce something that looks like an essay. It will tell you it’s great. And you might not find out it wasn’t until you’re staring at a waitlist or a ding in February.

If you want guidance that will actually push back on you, help you find your REAL stories, and give you school-specific advice that reflects how admissions committees actually think — that’s exactly what ApplicantLab is built to do. It’s not AI telling you what you want to hear. It’s a structured process, built from reviewing THOUSANDS of real applications, that forces you to do the thinking yourself — with expert guidance at every step.

You can start with a free trial and see exactly what I mean. No credit card required.

This is work worth doing. Don’t skip it.