The corporate version of this happens to be employees who work from home who can’t even fake their alleged illness long enough to log on to a team Zoom, cough and hack for 30 seconds and say “sorry gang, I feel too lousy to even be on camera”. They just send a generic “I’m under the weather” email.
I say- “Really? You can’t even fake a root canal or abdominal surgery? You think the common cold is going to get you out of a half hour Zoom call?”
Remember when folks had to write in tiny letters the formulae for various chemical compounds underneath their shirt sleeves and then pretend to be looking at their watches when they were scrutinizing their cheat sheets? Cheating took time, creativity, some acting chops back then!
So let them all “cheat” (if that’s what you want to call it) and let them suffer the consequences down the road when they are ill-prepared for the rigors of well-paying work and are left in the dust by those honest scholars who did not allow AI to atrophy their brain muscles. Just another example of survival of the fittest and the widening divide between the haves and have nots. Meh.
The thing is that I am the instructor and I don’t want to let them all cheat (which is what I do call it, because that’s what it is). A significant part of my job involves making sure people leave my class just a little bit more capable, informed, and skilled than they were when they started it, and if I adopt the attitude that it’s not my problem and cheaters will pay the consequence later, then I’m not really doing my job. It’s just that my job is becoming more impossible and exhausting every semester.
You can teach them academic “right” from “wrong” and the consequences of not creating the invaluable neural pathways developed by honest scholarship and then let go of what you can’t control. You can only control what you teach, not what students actually learn.
In some disciplines, you can probably use AI effectively to teach. I understand that it’s especially effective in some lines of work that depend on diagnostics or statistical modeling (whether in medicine, tech, or other fields), for example. In others, you can’t use it and never will. Mine is in the latter group. Using AI is cheating (fine, I’ll grant exceptions for spelling and grammar checks). There is no part of the work I do, the content I teach, or the work my students are supposed to produce that can be improved or even legitimately attempted with AI.
I disagree. The core problem with cheating isn’t that cheaters will ultimately fail. In many cases, they won’t. As a society, we reward wins more than “winning the right way.”
The real issue is what cheating does to everyone else. It displaces people who earned their outcomes honestly. And rather than letting cheaters truly face consequences, we often absorb those costs collectively. Think of industries that privatize their gains but push their losses onto taxpayers when things go wrong. Cheating works the same way. The honest person loses ground, and the rest of us end up subsidizing the gap.
That’s the actual harm. It’s a societal problem, not just a personal one.
No argument that “cheating” in the traditional sense has consequences, but I think we need to redefine this term in the age of AI. Focusing on academia, I believe the onus is on educators to embrace this new tool and teach students how to use it to deepen their learning and thinking, not blame students for outsmarting our now outdated teaching methods and assignments. We need to reinvent what and how we teach such that the use of AI is integral to desired learning outcomes. If AI currently can be used to mail in assignments and tests, the problem is with the assignments and tests, not the students. I’m sure we’ll eventually figure this out such that the term “cheating” no longer applies.
Many students who are cheating on assignments and tests are simply copying the question into AI, then copying the answer out of AI into their assignment or test… my daughter has observed many fellow students doing this in class. If they can’t copy and paste directly, they will use AI on their phones, and then type in the answer. You don’t consider this to be cheating? I am failing to understand what you are trying to say here, and why you are putting the word “cheating” in scare quotes.
I’m saying tests/assignments/teaching methods need to be completely reinvented such that they don’t look like they do today, where there are no paper/online question-and-answer formats; that’s now the old outdated way. In other words, we need a whole new (perhaps solely Socratic?) teaching paradigm where AI is integral to the learning outcome, not a way around it. There will be no copying and pasting possible in this new teaching world because there will be nothing to (or no reason to) copy and paste.
Just realized we had this very discussion a while back on another thread.
There are some subjects where a test that takes an hour or few is not a particularly good proxy of learning (e.g. visual arts, writing a research paper, doing a term project). Also, standardized tests are often targeted for test-specific preparation, rather than academic activity to improve the actual target of learning that the test attempts to proxy.
For some subjects in some situations, standardized testing may be the best way to measure, but that could mean that it is the least bad way to measure (and it could depend on the design of the test as well).
Regarding “unbiased”, a test could be biased for or against something other than what the test intends to proxy. Test-specific preparation is one example of such.
The UC system should receive kudos for openly admitting problems and bringing them to light. For the most part grade inflation is not as prevalent as private schools and they deal with cheating (whether through AI or regular) sternly. Here is the berkeley equivalent of students not doing well in a intro class even though it is for other reasons.
Yes. one of the profs in the reddit thread confirmed it was academic misconduct. He also made a point that the class should not be considered easy, but taken seriously as a university level 4 credit course.
This is also true. It is possible to add .claude and .cursor folders in the assignments with code which can alert when these tools are used. Though I am not sure this was the case here.