A whole new ballgame in cheating. Introducing ChatGPT

[downer]

... your university has to decide between losing the students or having inadequate standards, even if the university can and wishes to put in the resources.

Do you consider choosing the latter to be academic misconduct on the part of the university? Even if the trustees consider the former to be fiscal misconduct.

The median degree from an median university has always been meaningless.  I don't know why we pretend otherwise.

It does seem to make a difference for employers for white collar positions. But I expect that won't be the case forever, or that at least the employability gap between the median degree of the median college and a HS diploma will narrow - that employers will increasingly focus on other qualifications and experience instead of treating a college degree as a requirement. You can't have falling standards and increased tolerance for snowflakery forever on the one hand, and soaring costs on the other, and not expect changes.

Definitely.

For pandemc-era students who graduated high school 2020-2, and now Chat-GPT-era students who graduated college 2023 onwards, I can't imagine employers are going to especially impressed by a high school diploma or a college degree on its own. These qualifications are largely meaningless for many degrees.

But a US college degree was never a strong guarantee of competence or a clear set of skills in itself. Even if students learned some skills, they soon forgot them. It more indicated a certain capacity for persistence and acculturation. I was always rather puzzled why employers insisted on a degree. It didn't make a lot of sense to me. There are a lot of social and cultural factors that are chipping away at the general belief that a degree means a lot. Colleges face a challenge in demonstrating that they are selling a valuable product.

[marshwiggle]

... your university has to decide between losing the students or having inadequate standards, even if the university can and wishes to put in the resources.

Do you consider choosing the latter to be academic misconduct on the part of the university? Even if the trustees consider the former to be fiscal misconduct.

The median degree from an median university has always been meaningless.  I don't know why we pretend otherwise.

It does seem to make a difference for employers for white collar positions. But I expect that won't be the case forever, or that at least the employability gap between the median degree of the median college and a HS diploma will narrow - that employers will increasingly focus on other qualifications and experience instead of treating a college degree as a requirement. You can't have falling standards and increased tolerance for snowflakery forever on the one hand, and soaring costs on the other, and not expect changes.

Definitely.

For pandemc-era students who graduated high school 2020-2, and now Chat-GPT-era students who graduated college 2023 onwards, I can't imagine employers are going to especially impressed by a high school diploma or a college degree on its own. These qualifications are largely meaningless for many degrees.

But a US college degree was never a strong guarantee of competence or a clear set of skills in itself. Even if students learned some skills, they soon forgot them. It more indicated a certain capacity for persistence and acculturation. I was always rather puzzled why employers insisted on a degree. It didn't make a lot of sense to me. There are a lot of social and cultural factors that are chipping away at the general belief that a degree means a lot. Colleges face a challenge in demonstrating that they are selling a valuable product.

This isn't the case for professional degrees where there is a lot of specialized knowledge a job requires. Very few people would ever pick it up on their own.

[Stockmann]

But a US college degree was never a strong guarantee of competence or a clear set of skills in itself. Even if students learned some skills, they soon forgot them. It more indicated a certain capacity for persistence and acculturation. I was always rather puzzled why employers insisted on a degree. It didn't make a lot of sense to me. There are a lot of social and cultural factors that are chipping away at the general belief that a degree means a lot. Colleges face a challenge in demonstrating that they are selling a valuable product.

Talking to an Oxford student once, I was told that the reason UK employers, private and public, tend to strongly prefer Oxbridge graduates for high-flying positions over those from other universities wasn't mainly because of their knowledge or technical skills (though they were presumed to have gotten a good education, of course) nor because of their raw intelligence (though they were presumed not to be stupid) but rather because, due to the workload, tight deadlines and no-nonsense, no-excuses institutional attitude, they were able to work well under pressure, organize their time well, prioritize, work hard, etc. In other words, the claim was an Oxbridge degree is seen, among other things, as a certificate of non-snowflakery.

In a similar vein, I guess it was once a reasonable assumption a US college degree at least meant you were able to show up reliably and on time, were trainable, could manage deadlines, etc. It wasn't crazy, if you assume that's true, to make it a prerequsite for most white collar jobs, esp. as HS diplomas became little more than a glorified birth certificate.

[downer]

We could debate about whether the reality of Oxford matches up to the image, but the general idea makes sense.

I'm not necessarily advocating for a return to the "good old days." There's no turning back. But college, and especially the goal of nearly everyone going to college, may just not make any sense any more, if it ever did.

Looking at all the nicely formatted bibliographies referencing 6 scholarly articles that were not much discussed in the papers this semester, I certainly suspected that most of those papers were created with the help of AI. In previous years, it was a rare thing for students to be able to do that formatting, even with the help of citation machines. Turnitin flagged a few of them, but that's not proof. I am still processing, but it's pretty clear that the whole project of a college degree is being undermined and so far most of the purported solutions to this are themselves problematic.

[Caracal]

We could debate about whether the reality of Oxford matches up to the image, but the general idea makes sense.

I'm not necessarily advocating for a return to the "good old days." There's no turning back. But college, and especially the goal of nearly everyone going to college, may just not make any sense any more, if it ever did.

Looking at all the nicely formatted bibliographies referencing 6 scholarly articles that were not much discussed in the papers this semester, I certainly suspected that most of those papers were created with the help of AI. In previous years, it was a rare thing for students to be able to do that formatting, even with the help of citation machines. Turnitin flagged a few of them, but that's not proof. I am still processing, but it's pretty clear that the whole project of a college degree is being undermined and so far most of the purported solutions to this are themselves problematic.

I'm not sure what the problem is really. Who cares if the bibliographies are well formatted if they don't actually make any real use of the sources in their papers? There are some kinds of papers that might be particularly vulnerable to AI cheating, but those are mostly the kind that would have been vulnerable to other kinds of plagiarism. A little adjustment might be needed here and there on assignments and grading, and perhaps in some disciplines its a little tricker, but this isn't some sort of existential crisis.

[Hibush]

I'm not sure what the problem is really. Who cares if the bibliographies are well formatted if they don't actually make any real use of the sources in their papers? There are some kinds of papers that might be particularly vulnerable to AI cheating, but those are mostly the kind that would have been vulnerable to other kinds of plagiarism. A little adjustment might be needed here and there on assignments and grading, and perhaps in some disciplines its a little tricker, but this isn't some sort of existential crisis.

Lit searching has gotten very good. Both Google Scholar and Web of Science will provide the most influential papers on your topic of interest with a decent prompt. Importing those into citation software and formatting a bibliography is trivially easy.
That part is designed to be trivially easy because it facilitates scholarship. The scholarship is, as you say, in reading and analyzing those papers and writing a clear synthesis of that analysis in light of the question the paper is addressing.

AI is not very good at faking the scholarship. Perhaps the language of typical papers, and some superficial statements drawn from the resources. If anything the availability of these text generators puts more pressure on scholars to make their insights very clear and not to hide in the jargon of their subfield.

[downer]

Most students in intro classes are not very good at using the research. They still got a B in the past, other things being equal.
Now they use AI, which is not very good at using the research. Generally, it is hard to tell the difference between a paper done by a B student and work done by AI, except that if the bibliography is badly formatted, the student probably wrote it.

One solution is just to start to give C or D grades for papers that don't do well at integrating the research. It would be a paradoxical effect of AI -- everyone's grade goes down.

[Wahoo Redux]

CHE: Why I'm Excited About ChatGPT

Lower Deck:

Here are 10 ways ChatGPT will be a boon to first-year writing instruction, Jennie Young writes.

[Hibush]

Most students in intro classes are not very good at using the research. They still got a B in the past, other things being equal.
Now they use AI, which is not very good at using the research. Generally, it is hard to tell the difference between a paper done by a B student and work done by AI, except that if the bibliography is badly formatted, the student probably wrote it.

One solution is just to start to give C or D grades for papers that don't do well at integrating the research. It would be a paradoxical effect of AI -- everyone's grade goes down.

You could publish the ChatGPT response to the prompt and say that is a D paper. If they want a higher grade they have to demonstrate that they understand the literature and have thought about the question.

[downer]

CHE: Why I'm Excited About ChatGPT

Lower Deck:

Here are 10 ways ChatGPT will be a boon to first-year writing instruction, Jennie Young writes.

Reading through that, my first thought was "Is she on crack?"
I guess her fundamental assumption is that students are keen to learn, rather than being keen to do as little work as possible.

[downer]

Most students in intro classes are not very good at using the research. They still got a B in the past, other things being equal.
Now they use AI, which is not very good at using the research. Generally, it is hard to tell the difference between a paper done by a B student and work done by AI, except that if the bibliography is badly formatted, the student probably wrote it.

One solution is just to start to give C or D grades for papers that don't do well at integrating the research. It would be a paradoxical effect of AI -- everyone's grade goes down.

You could publish the ChatGPT response to the prompt and say that is a D paper. If they want a higher grade they have to demonstrate that they understand the literature and have thought about the question.

It's worth trying I guess. Has anyone tried this?

[apl68]

CHE: Why I'm Excited About ChatGPT

Lower Deck:

Here are 10 ways ChatGPT will be a boon to first-year writing instruction, Jennie Young writes.

Reading through that, my first thought was "Is she on crack?"
I guess her fundamental assumption is that students are keen to learn, rather than being keen to do as little work as possible.

In what sorts of ways does she think it would help?

[apl68]

Most students in intro classes are not very good at using the research. They still got a B in the past, other things being equal.
Now they use AI, which is not very good at using the research. Generally, it is hard to tell the difference between a paper done by a B student and work done by AI, except that if the bibliography is badly formatted, the student probably wrote it.

One solution is just to start to give C or D grades for papers that don't do well at integrating the research. It would be a paradoxical effect of AI -- everyone's grade goes down.

You could publish the ChatGPT response to the prompt and say that is a D paper. If they want a higher grade they have to demonstrate that they understand the literature and have thought about the question.

Now that might help.  Certainly now that systems like this are available, they have to be acknowledged and turned into teachable moments somehow.

I hope what you say above about this forcing scholars not to use so much boilerplate jargon comes true.  So much of what we read is so full of predictable jargon that it's already looked like the writers have been on autopilot for years.

[ciao_yall]

CHE: Why I'm Excited About ChatGPT

Lower Deck:

Here are 10 ways ChatGPT will be a boon to first-year writing instruction, Jennie Young writes.

Holy sheep.

[Stockmann]

I'm not necessarily advocating for a return to the "good old days." There's no turning back. But college, and especially the goal of nearly everyone going to college, may just not make any sense any more, if it ever did.... I am still processing, but it's pretty clear that the whole project of a college degree is being undermined and so far most of the purported solutions to this are themselves problematic.

Actually, I think it's the "everyone needs a college degree" approach that undermined the college degree, ChatGPT is just a receding tide that makes it glaringly obvious some boats were stuck in the sand all along*. Because the only way for everyone to get a college degree is to keep lowering standards until everyone does. Like it or not, an entire generation can't all be above average. That's how you go from teaching Latin in HS to teaching remedial English in college, that's how an A really comes to mean "average" and that's part of the reason for the present-day popularity of contempt for academia - at least at some level, many people realize a lot of college students are gaining little in cognitive abilities, hard or soft skills, but are paying through the nose for it.


*I bet ghost writing, plagiarism, etc have been common in college take-home essay-writing for a very long time, so courses relying heavily on them haven't really ensured learning for a long time - students in such courses who truly do the work probably benefit, but others get the grade while learning little or nothing.