Widespread AI cheating in college

[MU82]

Brown University professor Roberto Serrano suspected that students had used AI to cheat on a take-home midterm exam after the average score was 96; it's usually in the 60s to 80s. So he made the final in-person, and 27 students either dropped the class before the final or didn't show up to take it. Of those who did take it, the average score was 48.6, by far a historic low; three students scored 0.

https://www.insidehighered.com/news/faculty/learning-assessment/2026/07/08/brown-professor-suspects-most-his-class-used-ai-cheat

Three-quarters of Brown professors said they are concerned about students using AI to cheat, according to feedback from 105 faculty members.

The problem, of course, is not unique to Brown. The same share of respondents said the same in a 2025 nationwide survey from the American Association of Colleges and Universities.

Said Serrano:

"We cannot afford to have a society in which a significant fraction of our best young minds think that cheating is OK. That leads to a declining society, to a failed society. ... We cannot choose to become idiots."

[The Sultan]

Professors are going to have to learn to work with AI and understand that it's not going away. It's a tool to be used, with skepticism of course. But claiming it is "cheating" means they have a long way to go.

[JWags85]

Professors are going to have to learn to work with AI and understand that it's not going away. It's a tool to be used, with skepticism of course. But claiming it is "cheating" means they have a long way to go.

Yep, its another level in the "professors vs technology" battle that has been going on for DECADES.  I don't mean to rag on professors, but many are historically very slow to adapt or outright stubborn at the prospect of changing in response to technological advancements and progressions of their student environment.  My aunt was a non-tenured faculty Math instructor (not sure what the appropriate title was/is, she was a full time professor for decades with TAs, but not tenured nor did she do research).  She's been retired since right before COVID, but she's very much a luddite, didn't like laptops in her lectures even into the 2010s, and would be losing her mind about AI right now.  Also fully agree that "cheating" is a silly word to use here.  Either you administer a test in the classroom and allow no outside inputs or its "open book" and you accept that said book is now all the technology that students have access to.

I mean, 15 years ago, when I was in grad school, I had a course that had a weekly online quiz taken the day before class.  10-15 questions, obviously open book given that it was at home, but timed tight enough that you couldn't flip through the book to find every answer if you didn't read/prepare properly.  However, the textbook came with an online PDF version and nearly every question was a literal excerpt from the book.  Thus a simple Control + F to the PDF and voila.  I'm sure if those quizzes were done on paper in a classroom, the scores would have fallen through the floor as well.

Funny enough, for all the doom and gloom over AI in terms of academics, people are still unwilling or unable to do even the very minimum in terms of combating it.  You don't even need to use the AI checkers to recognize AI generated texts, posts, or papers.  95% of the people using them make no effort to tweak the output to sound more human or natural.  I'm in an industry WhatsApp group with about 50-75 other people that has a lot of chatter, news, and opinions in it, related to industry news, challenges, or other networking.  The proliferation of ChatGPT or equivalent word vomit shotgunned into the group by people is shocking.  It's very easy to recognize cause its always far more long winded than needed, overly formal, and sterile (especially coming from the large number of ESL members).

So as it relates to the topic, its only going to get worse (or better depending on what side you are on) so learning to coexist with it or what proper guardrails can be put on it are what is most important, cause complete banning or abstinence is unrealistic and putting a band-aid on a bullet wound.

[MU82]

Wondering if anyone knows what Marquette does to deal with it.

[Juan Anderson's Mixtape]

Brown vs The Bored of Education

[Billy Hoyle]

It's a major issue on campus. One of my old MU professors at another school now (I've been friends with him for 30 years now) has failed students who have used AI, turned in multiple students, and recommended suspension or expulsion for using AI when it's ridiculously obvious. He told me the most obvious one had different ideas in each paragraph, so it was obvious the kid just copied and pasted without even looking at what he was turning in.

My favorite story he's told me was pre-AI, when the plagiarism-detection software flagged a paper as "97% plagiarized." He emailed the entire class and offered the offending student (he didn't name anyone) a chance to come to him and explain themselves. The offending student did not come, but three others did!   

[Shaka Shart]

It's a major issue on campus. One of my old MU professors at another school now (I've been friends with him for 30 years now) has failed students who have used AI, turned in multiple students, and recommended suspension or expulsion for using AI when it's ridiculously obvious. He told me the most obvious one had different ideas in each paragraph, so it was obvious the kid just copied and pasted without even looking at what he was turning in.

My favorite story he's told me was pre-AI, when the plagiarism-detection software flagged a paper as "97% plagiarized." He emailed the entire class and offered the offending student (he didn't name anyone) a chance to come to him and explain themselves. The offending student did not come, but three others did!   

Not hard to find the predental scholar in that story

[21Jumpstreet]

Professors are going to have to learn to work with AI and understand that it's not going away. It's a tool to be used, with skepticism of course. But claiming it is "cheating" means they have a long way to go.

Perfectly stated

[MU Fan in Connecticut]

Yep, its another level in the "professors vs technology" battle that has been going on for DECADES.  I don't mean to rag on professors, but many are historically very slow to adapt or outright stubborn at the prospect of changing in response to technological advancements and progressions of their student environment.  My aunt was a non-tenured faculty Math instructor (not sure what the appropriate title was/is, she was a full time professor for decades with TAs, but not tenured nor did she do research).  She's been retired since right before COVID, but she's very much a luddite, didn't like laptops in her lectures even into the 2010s, and would be losing her mind about AI right now.  Also fully agree that "cheating" is a silly word to use here.  Either you administer a test in the classroom and allow no outside inputs or its "open book" and you accept that said book is now all the technology that students have access to.

I mean, 15 years ago, when I was in grad school, I had a course that had a weekly online quiz taken the day before class.  10-15 questions, obviously open book given that it was at home, but timed tight enough that you couldn't flip through the book to find every answer if you didn't read/prepare properly.  However, the textbook came with an online PDF version and nearly every question was a literal excerpt from the book.  Thus a simple Control + F to the PDF and voila.  I'm sure if those quizzes were done on paper in a classroom, the scores would have fallen through the floor as well.

Funny enough, for all the doom and gloom over AI in terms of academics, people are still unwilling or unable to do even the very minimum in terms of combating it.  You don't even need to use the AI checkers to recognize AI generated texts, posts, or papers.  95% of the people using them make no effort to tweak the output to sound more human or natural.  I'm in an industry WhatsApp group with about 50-75 other people that has a lot of chatter, news, and opinions in it, related to industry news, challenges, or other networking.  The proliferation of ChatGPT or equivalent word vomit shotgunned into the group by people is shocking.  It's very easy to recognize cause its always far more long winded than needed, overly formal, and sterile (especially coming from the large number of ESL members).

So as it relates to the topic, its only going to get worse (or better depending on what side you are on) so learning to coexist with it or what proper guardrails can be put on it are what is most important, cause complete banning or abstinence is unrealistic and putting a band-aid on a bullet wound.

My new US company head (I don't know what rock they found him under) is sickening with the amount of AI generated emails he sends to us.  There are only 5 other USA employees.  They are excessively and unnecessarily long in length and loaded with so much fluff we all want to barf.  The usual response is "dude you could have said that with 2 sentences.  We didn't need 3 paragraphs!"

[PointWarrior]

so just AI to summarize it, problem solved

My new US company head (I don't know what rock they found him under) is sickening with the amount of AI generated emails he sends to us.  There are only 5 other USA employees.  They are excessively and unnecessarily long in length and loaded with so much fluff we all want to barf.  The usual response is "dude you could have said that with 2 sentences.  We didn't need 3 paragraphs!"

[Jockey]

Modern educators are going to have to adapt - whatever form that takes.

But, yes, presenting someone's/something's words as your own is cheating.