AI in Higher Education: Are We Preparing Students to Lose to AI?

A Harvard research fellow just said something that should wake up every college, employer, and business owner in North America:

"What we're doing is preparing human beings to lose to AI — instead of preparing them with skills that complement AI." — Christopher Dede, Harvard Senior Research Fellow

Ouch. But he's right. And the same lesson applies to your business.

The Big Mistake: Fighting the Wrong Battle

Right now, most universities are asking one question: "How do we stop students from cheating with AI?"

Wrong question.

Professors are going back to paper exams and blue books. Harvard research says this "misses the point entirely." Why? Because banning AI doesn't prepare students for a world that already runs on it.

The reality check:

• 85% of college students already use AI at least every two weeks.
• Over half use it to write assignments.
• 70%+ of employers say they'd rather hire someone with less experience who understands AI than someone with more experience who doesn't.

The message is clear: AI isn't the enemy. Not knowing how to use it is.

Reckoning vs. Judgment: The Skill That Actually Matters

Here's the key idea. Harvard borrows two words to explain it:

Reckoning = fast calculation and prediction. Machines are amazing at this. So are standardized tests like the SAT and GRE. But guess what? So is AI.

Judgment = the human stuff. Weighing right and wrong. Reading context. Understanding people, pain, and quality of life. AI is terrible at this.

If schools only train students to "reckon," they're training them to compete with a machine that will always win. The future belongs to people who can judge — and use AI to handle the reckoning.

"Doing Things Better" vs. "Doing Better Things"

Most schools use AI to do old things faster — grade quizzes, auto-generate lectures. That's "doing things better."

The real win is "doing better things":

• Project-based learning where students solve real problems.
• Immersive simulations — imagine AI role-playing your tough boss, a used-car dealer, or a difficult client, so students can practise negotiation before the real thing.
• Collaboration that AI can't fake.

This is how you build judgment. Not by memorizing — by doing.

The 200-Applications Warning

Here's a story that says it all. A hiring manager posted one internship. She got 200 applications — nearly all identical, because students let AI write them.

She cut the AI-written ones first.

The lesson? When everyone uses the same tool the same lazy way, the people who stand out are the ones who add human judgment on top. That's the skill worth teaching.

Why This Matters Beyond the Classroom

Here's the part business owners can't ignore: your company has the exact same problem.

Buying Microsoft Copilot licences handles the "reckoning." But your team still needs judgment — knowing when to trust the output, spotting AI "hallucinations," and building AI into real workflows instead of just typing prompts.

Microsoft reported 10 million Copilot users in a single quarter. Adoption is exploding. But tools without training just create expensive confusion. The winners aren't the ones with AI — they're the ones who know how to use it well.

The Bottom Line

Higher education is at a crossroads — and so is your business. You can:

1. Ban and fear AI (and fall behind), or
2. Teach judgment and use AI to do better things (and win).

Harvard's prediction? "A lot of universities will change in order to survive." The same is true for businesses.

The question isn't if you adopt AI. It's whether your people know how to use it well.