I’ve recently seen several stories in the news about students who have “cheated” their way through college degrees using AI and other tools—even through the doctoral level. Interestingly, these students and some in society are celebrating this as an achievement. Educators, on the other hand, are terrified.

For as long as we’ve had education, teachers have fought against academic dishonesty in their students. It’s as if the students are fish, swimming around in a bowl where the water level is continuously rising closer to the rim, allowing the fish to easily jump out. As student-available technology pushes the water level higher, our educational systems and faculty build the walls higher and higher with technology and solutions of their own, trying to prevent the fish from escaping. This never-ending arms race has kept up for centuries, accelerating with each new tool and technology that has been developed—first the means of cheating, and then its antidote—sustaining the belief in hopeful faculty that they can “keep students honest” and force them to really learn. 

Suddenly, AI became widely available to the student masses, completely changing the landscape of teaching, learning and work. In other words, the fish spontaneously sprouted wings. Soon after, agentic AI created another leap forward in capability and ease-of-use. Forget the wings—these fish are wearing jetpacks. Faculty everywhere are scrambling in search of solutions that fit their paradigm of teaching and learning, from returning to paper-and-pencil work (tethering the jetpack-wearing fish to the bottom of the pool), to anti-AI software (building a roof over the pool). True panic is spreading among our colleagues as the prevailing models of teaching that have formed the tradition in college classrooms begin to crumble. There seems to be no real way to force students to “learn” in the way we’ve always been able to—especially if we want to keep using the tools that we’ve grown used to using.

I think we are asking the wrong question altogether. Instead of asking, “How can I keep the fish in the pool?” we should be asking “Why do the fish want to leave the pool, and how can we help them see the value of the water they are swimming in?” The answer, of course, requires a bit of context and framing. If we want to find a way to help students value real engagement with the learning process (swimming in the water), then we first need to understand the factors that are driving them to jump. I’ll give you a hint: I’m not about to launch into a discourse on the inherent dishonesty of students.

A colleague recently remarked at a faculty event that in their 20+ years teaching, they have never had such a high instance of cheating as they do now, as several others nodded in agreement. I don’t think they are wrong. Cheating—including the unethical use of AI—has become rather commonplace in academia at all levels—not just an epidemic, but a broad cultural shift in the student body, driven by a perfect storm of educational systems, teaching traditions, and cultural value shifts. The forces that created the situation we find ourselves in matter, because any true solution must address these causal factors.

Our educational systems have contributed to this situation in many ways, but for now I will focus on two. First—the Carnegie Unit, which has been around for about 120 years now, has become a near-universal measure of the amount of learning a student has completed. It is used to determine readiness for subsequent levels of education, sufficiency for degree attainment, eligibility for financial aid and athletics, and a host of other things. The problem? The C.U. equates “seat time” with a quantity of learning. This measure ignores virtually everything we know to be true about humans learn. The second educational system I’d like to address is grades. In addition to being verifiably bad for learning, grades are also non-standardized (an A in one Biology class is not equal to an A in another), only loosely connected to learning (hello, extra credit for attending the school basketball game), and subject to inflation over time (students today get much higher grades than they did 50 years ago). Put the CU and grades together, and you have a complex system where we focus the students’ attention on showing up for class, and jumping through the hoops to get the grade. Neither thing is directly connected to learning. Remember that for later.

Our teaching methods—particularly in higher education—can broadly be described as inauthentic and superficial. Specifically, when we favor rote memorization, multiple choice exams, and any learning activity that frames knowledge as a commodity or product—a substance that can be transferred from teacher to student—we are engaging in a type of teaching that rewards a product over a rich, engaged process. From an early age, we train students that we are going to tell them the right answers in the form of a lecture. Their job is to write them down, study them, and then parrot them back to us on an exam. In society, we routinely equate memory with learning. Admittedly, there are varying degrees of this, but a whole lot of what’s happening in higher ed classrooms fits under this umbrella. To summarize what we have so far, we are rewarding students for producing the right responses and sitting in a seat by giving them grades that everyone knows are actually a farce.

The third piece of the puzzle is the accelerant that really gets this fire going. It is the fact that we live in a culture when finding the path of least resistance is at peak value. Our societal fixation on the optimization of everything can probably be traced back to the industrial revolution, but in this age of free access to a universe of data, we seek to find the shortest path from A to B in every context possible. Tech companies “build the plane while flying it”, and sprint to the MVP (minimum viable product) so they can launch half-baked products as quickly as possible. “Gaming the system” is a celebrated victory, from political gerrymandering, to elite athletes finding new ways to skirt the rules in search of a win, to “lifehackers” and “lifestyle engineers” who obsessively search for shortcuts in everything from food to work to vacations. The line continues to blur between us and our shortcuts, as the work economy shifts toward further optimization using AI.

So putting this all together from a student’s perspective—we tell them “show up for class, and complete these tasks, and we will give you a degree, which you will need to get a job.” The tasks we give the students to do are often disconnected, rote, and repetitive, which is the origin of “When am I ever going to use this?” We warn students of a rapidly approaching job market in which everyone will need to know AI or be replaced by it. Everyone on the planet is hacking, optimizing, and gaming the system. Why on earth would a student value doing things the “hard way” when presented with this situation? They are simply optimizing their education in the same way the rest of the world optimizes literally everything else.

The solution I’m proposing is far from simple. First, educators need to understand that learning isn’t about right answers, or remembering content. Rather, learning is the construction of mental models that impact how we understand and interact with the world. This construction happens best when students engage with real problems, in real ways that are personally meaningful to them. Teachers need to understand the value of an authentic learning process, so that students can understand its value, too. We need to provide learning opportunities for students that feel real and meaningful. In a world of optimization and focus on the destination, we need to be able to sell the value of the journey itself. If we can do that, then we don’t need to fear technology, or cheating, or AI. We’ll have shifted the whole conversation around the value of education.