We already had AI proof education.
Then we did a university reform, partly with the excuse of aligning with the rest of the EU within the Bologna process (and I say "excuse" because that's what it was, because the politicians introduced some things with that pretense that weren't like that in the rest of the EU at all, and it was perfectly possible to comply with Bologna without doing them) and partly to copy the US/UK ways. And one of the pillars of that reform was continuous assessment, and evaluating coursework.
As a consequence of this, first of all working class students were royally screwed. Because suddenly it wasn't OK to just organize yourself to prepare the exam, you had to attend lots of sessions to earn points, which put students who work at a disadvantage. And second, passing by cheating became possible, even before LLMs. People tend to forget that before everyone got access to ChatGPT, some people had access to experts (family members, or even paying someone to do the work).
Now that this kind of cheating has been democratized and everyone can do it instead of just the most privileged with access to experts or money to pay them, people act all outraged. Although pretty much nothing is being done, except for using snake oil detectors, or sometimes increasing difficulty of assignments to make them LLM-proof (with which you screw the students who actually want to learn without LLMs).
They spent years indoctrinating us (professors) in training courses on how the old exam-based ways were wrong (the "Napoleonic" model, they called it... none of them seems to entertain the thought that maybe if it had been working essentially unchanged since Napoleon it wasn't that bad, and you need solid reasons to change it beyond "this is old so let's change") and the new ways were the bee's knees. Like in the Milgram experiment, it's difficult for people to back down and acknowledge that they have been wrong, even when the solution is obvious.
I definitely could tell the difference, though most of the time I just studied full 4-7 days before the exam.
So that was the justification used to switching to a less impactful final exam.
No idea how true that is.
We were also told learning a phonetic alphabet was better for young children learning to read than using the old ABC system.
As far as I have heard, that turned out to be based on one person's fantasy and zero evidence and has actually had negative impact on children learning to read.
To my knowledge they still teach about audio/visual/kinetic learners and how you should structure the way you learn around which one you are. This has been debunked for decades.
Not just the UK, pedagogy/education is a very soft science, along with any other field that revolves around human behavior (psychology, sociology, etc...).
Using AIs in experiments and studies will be an improvement even if they do not accurately reflect human behavior, just because you don't need a harm review and you can repeat your experiments multiple times under different variables.
It was easy to cheat on the assignments. Working on them in groups was common and sometimes encouraged. The only person you could really cheat was yourself (and a TA who had to grade one more exam)
In the UK it's common for exams to have almost all of the weight. In my Physics degree almost all courses were entirely dependent on the exams.
Including a final exam which examined the entire four-year MPhys course known as General Problems where even getting 30% was considered a good grade!
It is definitely a lot more work for the professors though, most of my family are teachers. It's a lot of assessments and it's very rare to have funding for TAs. Some think that the extra work with worthwhile for the sake of transmitting the knowledge more effectively, but not all of them do.
Frankly, you sound a bit bitter about it from the professor's perspective, and somewhat rationalizing why it is bad for the students. But students do generally appreciate it, and yes good students too, not just cheaters. I think both good and bad students end up learning more and hating the process less.
Your comments on Bologna do resonate though, it was very confusing when I continued to study in Germany and the Netherlands. The massive reforms were supposed to be for alignment with EU, but if anything it got more misaligned. They unified all 3 and 5 year degrees into 4 year degrees, but in most of EU all degrees are 3 years now, for instance.
Regarding the parent comment, indeed, my Computer Science degree was mostly hand-written exercises and exams, and it wasn't that long ago. The degree is about fundamentals, about understanding concepts and applying them, about the tools you need to learn anything in CS afterwards. You are expected to learn most of the practical skills for building software on your own, since they are ever-changing. And I have to say, that style of education has served me very well in my career.
PS: I was also surprised to learn that most of the undergrad exams in Germany, and some in NL, are oral. I can see how that might be a disadvantage to some, but writing is also a disadvantage to others. I quite liked it, less intense than a long written exam, and I think the professor can get a much clearer understanding of the student's grasp of the subject. But again, it's a ton of work for the professor, 20-30 mins per student one-on-one, giving them your full attention, adds up quickly.
Not sure when this was supposed to be the case, but for actual universities (not meant in a deragotary way, Germany has two types of higher eduction) in hard sciences, most classes are graded on a single written exam. Both in undergrad/bachelors and masters.
Unless things have drastically changed in the last five years...
I know that an oral exam might seem less serious and rigorous, but I do think the professor can get a better grasp of how much the student actually understands the subject through an interactive interview.
The best was when she barely unscrewed one of this big DIN connectors so at quick glance it looked fine, but wasn’t fully connected.
That's evil haha. It's the case where you unplug and plug again everything, changing seemingly nothing, but then it works
If it is mostly a ”show your work”/”show your reasoning” kind of grading where your width and depth of attempts are more important than success then it seems OK.
sounds like some of the technical exams i'ev taken, and/or one or two job interviews
Lots of skills from those old days that have been lost/ignored in the pretence of productivity.
The internet enabled all the complexity we have today. LLMs will have a similar effect, but instead of engineers actually having to understand the system (even in it's complexity) they will just be querying the oracle to build things or solve problems.
When the oracle can't help (or maybe refuses to) is when it gets interesting.
How do you define "productive?" Lines of code written per day? Bugs fixed per man hour? Fewest reported bugs per end user?
The fastest compiler in the world won't help you find all the runtime bugs that simply wouldn't have existed in the days of punch cards, when code was written with with more care and attentiveness since there wasn't a fast edit/compile/test development loop. YMMV, of course.
It's a shame that they are also way more susceptible to cheating with AI.
Assignments and projects are great for learning, but suck for evaluation.
Another example, lit classes where the grade is based on time limited, open book exams, hand written in "blue books"
Read the book, pay attention in class, spend 90 min writing an essay, and you are done.
However I suspect that there are many who 1) are more concerned about the short term outcome, 2) consider the degree/diploma to be little more than a meal ticket or arbitrary gatekeeping without any connection to learning, 3) view the work as a pointless barrier to being handed said diploma, and/or 4) don't see the value of human learning in a world where jobs are done by AI and AI systems routinely outperform humans on complex tasks.
A lot of Gen Z are ferociously anti-AI, but for tribal and emotional reasons, not because of a nuanced understanding - which is ironic, because the nuanced reasons for being wary of AI are much stronger than the usual talking points about "stealing art".
Being tribal and emotional is going to make Gen Z easier to replace, because nuanced strategic insight is less common and more useful.
The other thing that feedback feeds into is credentials. I realize that some people are dismissive of this aspect of the degree, but it is important to pursue further studies or secure a job. While you can argue that these people are only cheating themselves, and some of them are cheating themselves, a great many will continue to cheat as they advance in academia or the workforce. In other words, they are cheating others out of opportunities.
Personally, I dropped out despite a full ride+ becuase why would I put in work for a no name state school when I already has an FTE job as a developer out of high school anyway.
Turns out fraudulent action can still get the bag.
And for most students that’s all they really care about.
If the companies stop valuing the diplomas, students will stop paying tuition to attend, and the universities eventually collapse.
You can imagine a world where the Corporate or State AI handles education, tailors it to individual student levels and talents, and assigns work based on its own direct experience of a lifetime of interaction.
You can also imagine that in that world where most humans would be redundant - unless the AI was optimising for human-to-human jobs and for evidence of unusual insight and creativity, not managing bullshit work for corporate profit.
So a student who only understands the basics should be able to answer most of the easy questions and students who have a deeper understanding can answer the harder ones.
Well-written exams should feel pretty fair and leave students feeling like the result they got is proportional to the effort they put into studying the material (or at least how well they personally felt they understood the material).
Is this kind of test - many short questions - a standard thing for math in your country?
My university exams were pretty much all "2-question", in 90 minutes.
The first half was an essay where you have to reproduce a lesson from the curriculum, in your own words.
The second half was "the formulas" - you have to develop one or two formulas from first principles.
I once got an A- even though I got "the formulas" half very wrong. As the teacher explained later, I simply chose the coordinate system beginning at not the same place the textbook did. And this was supposed to be a bad teacher - he actually gave Ds to almost all of us (180 people). This was a makeup exam.
You've never been a teacher.
They were more prone to cheating before AI, too.
Cheating has always existed at some level, but from talking to my couple of friends who teach undergrad level courses the attitudes of students toward cheating have been changing even before AI was everywhere. They would complain about cohorts coming through where cheating was obvious and rampant, combined with administrations who started going soft on cheating because they didn’t want to lose (paying) students.
AI has taken it further, with students justifying it not as cheating but as using tools at their disposal.
I was talking to my friend about this last week and he was frustrated that several of his students had submitted papers that had all the signs of ChatGPT output, so he asked them simple questions about their papers. Most of them “couldn’t remember” what they wrote about.
It’s strange to me because when I went to college getting caught cheating was a big problem that resulted in students getting put on probationary watch and being legitimately scared of the consequences. Now at many schools cheating is routine and students push the boundaries of what they can get their classes to accept because they have no fear of any punishment. YMMV depending on the institution
IBM used to hire software developers based on aptitude test scores regardless of formal education, then put them through an extensive internal training program. It worked fine.
People who got through via cheating in college tend to be low performers in work for the exact same reasons.
An interesting side effect of the AI gold rush is that companies are starting to look critically at these do-nothing email jobs where someone forwards emails around and makes slides and Notion pages.
I’ve worked with many who occupied jobs that didn’t contribute much other than organizing text and sharing it around, but they got a pass because it looked helpful enough. Now it’s a lot harder to justify those positions when management realizes that having the not-really-competent person summarize communications and documents isn’t much better than having ChatGPT do it.
Unfortunately a lot aren't, they feel like they have to be there or these courses are the only path for them to get a good job. And unfortunately they end up in the workforce, too. You'll often see teams with one good developer and a lot of hangers-on.
Writing papers is a useful skill to have. And many students aren't very good at that. I taught some classes during my Ph. D. and supervised some students with their master thesis and PhD thesis work. Many students get their degrees without that really getting addressed. At least Computer science degrees in the Netherlands just spend very little time on writing skills. You get students with high school levels of English and Dutch and that's it.
I learned to write properly only when I started my Ph. D. My supervisor made me do it right before he allowed me to submit papers for publication.
AI might actually be good for education long term. It will result in a more personalized approach, which I think is good. There are plenty of ways to test students that are more engaging and interesting for both teachers and students than some of the old ways. You can't fake knowledge when you do a verbal test. Or test people with a good old written exam.
And of course for teachers, you can automate a lot of the verification work. This can be a lot of work.
But there were already heaps of problems with tech in education before AI.
My CS projects were often pretty free-form so in theory I could've just used AI - today, anyway. But a big part of the grade was a face to face interview where you actually had to talk about the code you wrote. Anyone lifting along with other people who didn't actually do any work would fall through then.
I could easily imagine a CS theory course that doesn't involve any programming language at all.
Which strikes me as a terrible way to teach and test programming skills. If you're teaching to program without so much as syntax highlighting, you're not preparing your students for anything that even remotely resembles the industry they aspire to work in.
Honestly, these days universities should probably find a way to incorporate AI into their teaching, rather than fight it. Anything else is betting that AI will not stick around, which strikes me as a hopelessly naïve bet. Especially for software development.
I don't pretend to have all the answers, I don't know how to teach systems thinking in a appropriate way either. But I'm pretty sure typewriters isn't it, unless your students are hoping to get hired by Ada Lovelace, it's just not going to be relevant.
Education isn't any churning out a cog fitted to a bigger gear. A college education should not be preparing you for intellij or familiarizing you with va code.
It should be teaching foundational skills with the discipline. And I just don't understand how you look at the amazing things we've accomplished since ww2 and say "the education system that taught this fucking sucks"
Students today aren't studying to repeat the progress made since the forties, that's been done by their grandparents, they're looking to drive the next 40 years of innovation, and if that involves typewriters I'll eat my hat
The vast majority of students and employers treat them as vocational certificates in practice, and the profession would almost certainly benefit from adapting curricula to more closely match that reality.
Foundational concepts are still necessary, but I don't buy the argument that we should continue teaching like it's 1946.
But yeah, everything was hand-written. On sheets of paper with pencil. I even had to write x86 assembly out by hand for my CPU architecture class. Of course, laptops were available back then but not cellphones and certainly not LLMs, so cheating by electronic means probably presents a stickier wicket now than it did back then.
The only exception is that when I got into grad level classes we did have some big programming projects. But most of that programming happened on sparc stations, and it was actually just easier and more productive to sit at the machine in person with its nice big (at the time) display with all the other folks doing programming projects. Those machines had the standard dev toolchains provisioned that weren’t easy (at the time) to do on a dorm room Mac or windows computer.
I really think a lot of the ways we can reduce reliance on AI for thinking is to just set up systems where it’s not an inviting or rewarding option.
I don't want to be polemic, but I really miss those days.
One thing I recall is that the grading policy made it very clear that minor syntax issues were inconsequential in handwritten answers. And more advanced classes only wanted pseudocode. Which are exactly the right priorities.
He concluded the class by talking about the importance of observing patients, and pointed out that he had tasted a different finger than the one he had put in the beaker.
If my college is doing this, I cannot imagine how many others are also impeding on their entire goal: education.
I'm sure they had some kind of submit your code as assignment and using testing as a way to grade the assignments.
Apparently you learn to double check your work!