Imagine a professor asking artificial intelligence to draft a syllabus, prepare lecture slides, generate discussion questions, and develop a grading rubric. Then imagine that professor opening the first class with a warning: students who use AI to complete their assignments will face academic misconduct charges.
The irony practically writes itself. Perhaps the professor could ask ChatGPT to explain it.
There are legitimate reasons to restrict AI in certain assignments. Students need to develop knowledge, practice reasoning, and demonstrate abilities they can exercise independently. But when faculty embrace AI as a professional convenience while treating student use as inherently dishonest, they owe their students something more persuasive than “because I said so.”
The familiar defense is that professors already possess the expertise students are trying to acquire. A historian can evaluate an AI-generated account because years of study have supplied the necessary judgment. A student encountering the subject for the first time may accept confident nonsense. Similarly, an engineering professor should recognize whether a generated solution violates physical principles that a novice has barely begun to understand.
That distinction matters. It does not settle the argument.
Expertise is a reason to verify AI output, not an exemption from intellectual responsibility. A professor who distributes inaccurate slides, assigns fabricated references, or delivers explanations they have not checked cannot defend the result by pointing to a doctorate. Credentials do not review lecture notes. People do.
Nor does professional status make every use educationally sound. Preparing a course is itself an intellectual exercise. Choosing examples, anticipating misconceptions, connecting concepts, and deciding what deserves classroom time are central acts of teaching. Outsourcing those decisions without careful engagement can weaken instruction just as outsourcing an assignment can weaken learning. Students can reasonably ask: if struggling through the work is essential to developing our understanding, does preparing the work contribute nothing to yours?
The issue becomes sharper when faculty describe their own AI use as efficiency and students’ use as laziness. Both groups face deadlines. Both juggle competing responsibilities. Both can use technology thoughtfully or carelessly. The difference between assistance and substitution must rest on what the person actually does, not which side of the lectern they occupy.
Consider feedback. A professor may object that students submitting generated essays have deprived themselves of practice. Fair enough. But students receiving generic, automated comments may likewise feel deprived of the attention their work deserves. A course cannot credibly demand authentic intellectual engagement from students while offering them an automated approximation of engagement in return.
This does not mean professors must personally compose every quiz question or that students should be free to submit machine-written work as their own. It means the rules need an educational justification that applies consistently.
If an assignment measures unaided writing, prohibit AI drafting and explain why. If it measures the ability to evaluate competing arguments, consider allowing AI suggestions while requiring students to identify weaknesses, verify evidence, and defend their conclusions. If a calculation tests foundational understanding, require independent work. If a project tests engineering judgment, assess assumptions, validation, and the student’s ability to explain the design.
Faculty should practice comparable transparency. Tell students when AI substantially contributed to course materials. Check every factual claim and reference. Take responsibility for errors. Do not upload student work into external systems without appropriate authorization and privacy protections. Above all, demonstrate the habits you expect students to develop.
Assessment also needs to change. Asking students to produce an answer that a chatbot can generate instantly, then relying on threats to preserve the assignment’s value, is a fragile teaching strategy. Oral explanations, staged submissions, classroom exercises, and questions tied to specific project decisions can make understanding visible. These approaches require effort. So does meaningful teaching.
Universities should resist turning this debate into a contest between technologically enlightened faculty and supposedly untrustworthy students. Students deserve neither blanket suspicion nor permission to bypass their education. Professors deserve useful tools, accompanied by the obligation to use them responsibly.
The strongest classroom policy begins with a shared principle: everyone must remain accountable for the thinking behind the work bearing their name. A professor can use AI and still teach seriously. A student can use AI and still learn deeply. Either can also use it to conceal intellectual absence.
The real embarrassment is not that professors use the same tools as their students. It is that some expect students to recognize a distinction their own teaching policies have never bothered to explain. Academic authority carries greater weight when it can survive the standards it imposes on others.