In an AI Powered World, It's Time to Make Education More Human Than Ever Before

The first time one of my college students turned in an essay written by ChatGPT, I confronted them and threatened to evoke the college’s academic integrity policy.

The student denied it.

But as a technology enthusiast, I was extremely familiar with the nature of ChatGPT’s output. And, as a college instructor, I was also a frequent reader of student work. The two types of writing are distinctly different, and the use of AI, in this case, was glaringly obvious.

When I pasted the essay into ChatGPT and asked if it was AI-generated, the bot told me that it was “highly probable.” But when I pressed the AI to reveal its rationale, it told me that the essay had “complete sentences, proper grammar, and no spelling mistakes.” In other words, ChatGPT thought the essay was AI-generated simply because it was well-written.

In that moment, I knew educators everywhere were in for a bumpy ride. The classroom experience would have to change. And quickly.

Over the past two years, I have experimented with different ways of working in the classroom. I have thought deeply about the reasons why students would (and would not) want to use generative AI (GenAI) in their learning experiences. What follows are some of my recommendations for teaching and learning in an AI-powered world.

First, we must abolish traditional grades. This will be a difficult pill to swallow for many.

Yet, the most common reason students would turn to a generative AI tool to do their work is that they want to get a “good grade.” Specifically, an A. Experts such as Alfie Kohn, Robert Talbert, Susan Blum, and Laura Gibbs have pointed out that grades act as extrinsic motivators that actually hinder genuine learning motivation. A growing “ungrading” movement has emerged among educators, who are exploring assessment practices that don’t rely on traditional grading systems.

Grading systems that turn educational institutions into a video game, rewarding students with meaningless points, make it very tempting to game the system using AI. Educators can remove the focus on getting the “best” grade by allowing for errors, mistakes, and misunderstandings and permitting students to redo work that doesn’t meet the standard. Additionally, thorough written feedback and well-designed rubrics can point students in the right direction for academic success.

Second, we need to teach students the fundamental differences between themselves and chatbots that can seem so convincingly human. While removing grade-based incentives reduces the temptation to use AI deceptively, understanding AI’s limitations is crucial for students to appreciate their own unique capabilities. It’s critical that students understand AI not as a magic box that produces flawless essays, but as a sophisticated tool with significant limitations.

After all, ChatGPT has never had its heart broken. Claude has never felt the sun on its back while walking on the beach. Google Gemini has never enjoyed the taste of ice cream on a hot summer day.

These experiences fuel our creativity, empathy, and critical thinking—irreplaceable human skills that AI cannot replicate, and that the marketplace will crave as chatbots become more ubiquitous. By focusing on nurturing these qualities, we prepare students not just to coexist with AI, but to excel in ways that are uniquely human.