Professor Gives Everyone Full Marks After AI Takes Over His Classroom

A college professor spent weeks reading final papers that barely sounded like the students who submitted them. Then he made a decision that left the entire class with the same grade: 10/10.

The reason was not generosity. He had reached the point where figuring out which papers were genuinely written by students had become almost as exhausting as teaching the course itself.

The Professor Had One Brutal Realization

“I hate, absolutely hate how AI has forced me to turn into a punitive detective, rather than, well, a teacher.”

That was the opening salvo in a message sent to a college class after the professor reviewed their final papers. The professor has never been publicly identified, and the message circulated through screenshots after a student shared it online.

What he found left him deeply disappointed.

He described seeing “a level of AI-generated writing” that had become so extensive that many submissions no longer reflected the students’ voices, thinking, or the skills the course was supposed to develop.

The professor’s frustration was not simply about students breaking a rule.

He believed they were bypassing the very work they had enrolled to learn.

The Assignment That Became Pointless

One student had reportedly spent three weeks writing more than 3,000 words arguing against the use of AI, including its use in education.

They said they were proud of the paper and had hoped the professor’s feedback would help them become a better writer.

Instead, the assignment became part of a much larger problem.

The student’s Reddit post, shared on December 14, 2025, described the frustration of putting weeks of effort into work while classmates allegedly relied on AI. The post attracted more than 24,000 views.

One commenter identified themselves as a college writing instructor and wrote: “I teach college writing and I felt this in my bones.”

Then He Gave Everyone A Perfect Score

The professor’s response was almost impossible to square with what came before it.

“Given the extent of AI-derived work I have encountered, I am making an unusual decision for the Position Paper assignment: I am giving everyone a 10/10 on the final paper.”

He also cancelled the next assignment.

“I have also decided to cancel your Media Analysis assignment. You do not need to submit anything and everyone will receive full credit.”

That could sound like the professor had simply given up.

His message made clear that was not how he viewed it.

“This is not a reward. It is a warning.”

He argued that students who left the course without practicing their writing would eventually face consequences academically, professionally, and personally.

Then came the line that has become the defining quote from the entire episode:

“I cannot want your learning more than you do.”

Why The Professor Stopped Playing Detective

The professor’s decision was partly about something more complicated than cheating.

He no longer believed that trying to distinguish human writing from AI writing was serving the purpose of his course.

“At this point, attempting to separate authentic work from AI-produced work has become counterproductive, and this course was never meant to be a surveillance exercise,” he wrote.

That creates a difficult problem for universities.

A professor can suspect that a piece of writing was generated by AI, but proving it is another matter. Automated detection tools also have documented limitations.

A Stanford research team tested several GPT detectors using writing from native and non-native English speakers. The study found that the systems consistently misclassified non-native English writing as AI-generated, while native writing was identified more accurately.

That creates an uncomfortable possibility: a student could produce their own work and still be accused of using AI because their writing triggers a detector.

The professor therefore faced two competing problems:

• AI-generated submissions: Students can hand in work that does not represent their own writing or reasoning.
• False accusations: Detection systems can incorrectly flag genuine student writing.
• Teacher workload: Instructors can end up spending time investigating authorship instead of teaching.
• The purpose of grading: A perfect score becomes difficult to interpret when the work itself may not represent the student’s ability.

The result is a classroom where the grade can become less informative than it used to be.

AI Use Among Students Is Already Widespread

The professor’s experience also fits into a broader shift in student behavior.

A 2026 Higher Education Policy Institute survey of 1,054 full-time UK undergraduates found that 95% had used generative AI in at least one way.

The survey found 94% had used it to help with assessed work.

There was also a smaller but significant group using AI text more directly.

According to the survey, 12% reported putting AI-generated text straight into assessed work. That figure had risen from 8% in 2025 and 3% in 2024.

The numbers suggest that students are not all using AI in the same way.

Some may use it to brainstorm ideas, find information, organize thoughts, or improve a draft. Others are allowing the technology to produce material that ultimately gets submitted as their own.

That distinction is becoming one of the hardest questions for educators to answer.

The Bigger Concern Is What Students Lose By Outsourcing Writing

The professor’s warning went beyond essays.

“When you outsource that work, you weaken not just your ability to write, but also your ability to reason, to communicate, and to advocate for yourselves.”

That claim has some experimental evidence behind it, although the research is still limited.

An MIT Media Lab study put 54 people into three groups while they completed essay-writing sessions. One group used ChatGPT, another used a search engine, and the third used neither.

The researchers recorded participants’ brain activity using EEG.

The ChatGPT group showed reduced brain connectivity compared with the other groups, with connectivity scaling down alongside the amount of external support used.

The study also reported that ChatGPT users performed worse when asked to quote from essays they had written only minutes earlier. Participants using the chatbot also reported lower feelings of ownership over their work.

The findings are not enough to establish what years of AI-assisted writing will do to students.

The study involved a small sample, short writing sessions, and timed prompts. It was also a preprint awaiting peer review.

But it raises a question that the professor’s message makes difficult to ignore: if students repeatedly outsource the hardest part of thinking, when do they actually practice thinking?

One University Tried A Different Approach

Not every educator is treating AI as something that must be eliminated from student work.

Researchers Brian Harfe and Aaron Thomas at the University of Florida studied an assignment where students began with an AI-generated draft and then revised it into their own argument.

They tracked authorship across 310 essays.

Students retained roughly 76% of the machine-generated wording. Only two of the 310 submissions were returned without changes.

That experiment points toward a different classroom model.

Instead of pretending AI does not exist, educators can make students confront the weaknesses in machine-generated writing and then require them to reshape it.

It does not settle the larger debate, but it shows that AI can be incorporated into an assignment without automatically becoming the student’s substitute.

The Student Who Opposed AI Got The Same Grade

There is an especially strange twist to the professor’s decision.

The student who reportedly spent three weeks writing more than 3,000 words criticizing AI received the same 10/10 as students whose work the professor believed contained extensive AI-generated material.

On paper, they had identical results.

The work behind those grades was anything but identical.

The professor’s decision therefore stripped the grade of its usual meaning. The 10/10 was no longer a judgment that every student had demonstrated mastery.

It was a signal that the assignment itself had stopped giving the professor reliable information.

And there was still one route left for students who wanted genuine feedback.

They could email their own writing to the professor.

He would still read it.

He would still respond.

That may be the most telling detail in the entire message.

The Real Test May Be What Students Can Still Do Alone

AI has made it easier to produce polished writing without necessarily developing the skills required to produce that writing independently.

That creates a problem that no detector can fully solve.

A student can receive a perfect grade, submit an impressive essay, and still walk away without knowing whether they could have written it themselves.

The professor’s final message cuts through all of that: “I cannot want your learning more than you do.”

For a classroom increasingly filled with machine-generated words, the most valuable thing left may be the ability to put your own words on the page.