THE FUTURE OF EDUCATION IN AN AI WORLD

AI is changing what learning looks like, and it’s forcing a real question: If anyone can generate a correct answer in seconds, what should we as students actually be graded on?

That question has been on my mind for a while, and recently it became the focus of a thesis project I completed for my MAN 336H class within the McCombs School of Business. Through interviews with students, professors, and policy leaders, I started to see a pattern: the issue isn’t just AI use, it’s the structure of an education system built almost entirely on the final outcome.

We reward the finished essay, the final answer, the clean submission. But AI makes the outcome the easiest part.

Growing up in a classical school (Providence Classical School) helped me see the value of the opposite approach. We spent a lot of time learning how to reason and defend reasoning than learning just what to think. One assignment I still remember involved true-or-false questions where your explanation mattered more than the final selection. If you justified your reasoning well, you earned points. The teacher wasn’t evaluating the bubble you filled in, they were evaluating our thinking.

That idea matters a lot more in an AI world.

Process vs. Outcome

In my research, I noticed that most of the confusion around AI comes from trying to enforce outcome-based grading in a world where outcomes can be generated instantly.

Students feel unsure about what’s allowed. 

Professors keep adjusting rules. 

Administrators send new AI memos every semester.

Everyone is reacting to the same thing: The current system isn’t built for a world where “the answer” isn’t proof of learning.

What is still a signal of learning is the part AI can’t do for you:

• the reasoning

• the decision-making

• the judgement

• the ability to compare options

• the ability to defend your thinking

• the strategy behind using AI or choosing not to

That’s what your process reveals.

A Dual-Grading Model Makes More Sense

I believe education will move toward a dual-grade system:

1. A Process Grade

How you think, research, organize, iterate, and use AI intentionally. This rewards the path you took. Not whether you used AI, but how.

2. An Outcome Grade

The quality of the final product.

Together, this gives teachers a more complete picture of actual learning. And it does something important: It removes the shame and secrecy around AI use while still holding students accountable for their thinking.

Mindless use → lower process grade. 

Strategic use → higher process grade. 

No use → still fine, as long as the reasoning is strong.

This is much healthier than trying to detect, punish, or restrict, which creates anxiety on both sides. The structure becomes aligned with reality instead of fighting it.

Where This Is Heading

We’re already seeing early versions of this shift:

• classes that require draft histories

• assignments built around reflection and analysis

• teachers grading “show your work” more than the solution

AI isn’t ruining education. It exposed what learning was always supposed to be. 

Closing Thought

If education adapts around reasoning, judgement, and intentional use of technology, students will end up stronger thinkers, not weaker ones. The goal isn’t to eliminate AI or encourage dependence on it, but to design structures where students are incentivized to use it with clarity, discipline, and strategy.

That’s the future I see emerging.

Of course, a dual-grading system isn’t without challenges. A clear one is that it can potentially increase the grading load for faculty.

Teachers would need a way to quickly review a student’s reasoning, drafts, choices, and interactions with AI. Without support, that becomes overwhelming.

That’s why any future version of this model would need to be paired with technology that assists, not replaces, the educator’s judgement.

Imagine an AI tool that professors can customize to their preferences. It could analyze a student’s process notes, drafts, and decision patterns and generate an initial rubric-based assessment of the “process grade.” But the professor would still have final say, reviewing flagged sections, overriding decisions, or asking a student to clarify their reasoning.

Students could even have the opportunity to defend why they believe their process deserves a different evaluation, a skill that mirrors real-world communication and decision-making.

If you want to read my full thesis: including the interviews, frameworks, and policy breakdown you can read it here: → [Read the full thesis]