Make them write out the pseudo code by hand for thier projects? from what I can gather, the issue is that they don’t understand the underlying code that the Ai created? Thus bug testing or altering values manually is not in thier wheel house?
I’m not an expert on this matter but I don’t hear anything about pseudo code, which allows you to write code for anything… Just got to translate it into code/a language that the computer can use?
This has been an issue with schools since forever. Focusing on grades instead of learning means many kids will do what is needed to get a good grade, pass the class and get the degree.
When we stop valuing grades and start valuing knowledge directly, learning improves. Knowledge should be the reward and what we laud. The grade is just a mechanic for evaluation, but should be considered no value in itself. That’s what kids need to be taught and shown from day one.
When your kid comes home from school, ask them what they learned. Ask them to teach you as if you don’t know, or maybe you really don’t. Don’t correct them, instead ask questions that might lead them to the correct answer, if you know it. People learn best by teaching others. When you give your kid the expectation that they will be teaching you what they learned, even casually, they pay attention in class and do much better.
It’s much better than doomscrolling tiktok for both of you, trust me.
Grades are the measure of kids are progress in what learning and what knowledge they actual have, you can call it something else like in Montessori where they change the system of grading where they use terms like mastery but they still have a system in place to track progress.
I think you’re confusing grades with teaching to the test, thanks No Child Left Behind btw, here which okay is a very valid point but end of they day standardized tests are a way to measure your actual grasp of the material being taught. Sure, how you get there needs adjustment at least in America where our math scores and reading comprehension scores are abysmal but to say it’s grades is the issues is kind of reductive when it’s such a more complex issue that involves how we teach, how we deal with disparate incomes, new thing is low enrollment, and why the hell aren’t we teaching phonics when whole-language and three-cueing methods are proven to be ineffective for literally decades now.
There’s a reason why colleges are moving back to requiring ACT/SAT scores in their applications again because in particular with items like math, kids were graduating and then finding out that they didn’t understand basic concepts like Algebra.
Also, this conversation has nothing to do with the article and AI.
you can’t measure knowledge though.
schools teachers and students are measured, they are judged by numerical performance metrics.
also, after grammar school parents have no clue what their kids are doing or learning. most adults can’t do algebra or read beyond a 7th grade level.
plus what you are implying would require massively more teachers and time in education, something nobody wants.
Who the fuck enrolls in an educational course, if they wish to skip learning?
Increasingly higher ed in america is being treated as a degree mill, where students are “customers” and the degree is a “product.”
It’s disgusting, but it’s a natural consequence of a commodified education system in which colleges and universities are treated as businesses.
I personally think we need to cut out the parasitic administrators and let the department faculty run the schools like co-ops, and possibly hire out the few administrative tasks that are actually necessary.
That would be far cheaper than paying these enormous administrative salaries, so it could lower tuition while increasing faculty pay.
But has anyone considered those poor shareholders?
most people hate learning. because it’s difficult and requires effort and time.
cheating is easy. you get results with less time or effort.
As all the others say, they want the cert and the status that comes with the cert.
It’s part of the reason everyone needs three years experience for an entry level job, a good student fresh out of education is worth little but then worth a lot if they learn fast. A bad student fresh out of education won’t learn at all and will be an active detriment, if only because they’re taking a seat someone better could take.
People who use AI for everything aren’t guaranteed to be bad students, but bad students probably will use AI for everything. The trick will be figuring out the difference.
Oh, you’d be surprised.
I used to teach english for non native speakers, and after chatgpt’s debut in 2023, almost every single essay, homework, and presentation was entirely done with LLMs. We had to demand handwritten essays, and presentations had to be done in the classroom (we gave the theme, they worked out a presentation, and then presented it the same day without leaving the class).
And even then, some students would ask to go to the bathroom, use their phones, memorize a whole ass paragraph from chatgpt and then come inside class to write it down.
The mentally bankrupt
So i imagine if a student doing a video game course is doing it because they are passionate about gaming, and these students are willing to use AI to shortcut something they actively want to do, then its not a big leap to assume they are using it to deal with all the other subjects they find tedious.
I feel like we are heading for a future of flesh robots interfacing with the internet to steer their lives, and the second the internet goes down people will be wandering around aimlessly, looking dazed in an internal conflict between wondering where to start and being confused that the black glass in their hand cant give them answers.
ugh,we’re so fucked
Stupid in, stupid out applies to teaching as well. 🥲
My two cents on this is that the education system is lagging.
Implied, thing’s are changing quickly too soon for any education to happen. A lot of education platforms are training companies llm modules. Students are adapting by using llms but it doesn’t produce more data. All big tech that’s invested in llms are pushing it harder down your throat to the point of canceling contracts as threats.
And then there’s always the “which studio will be shut down today?”
On one hand employers are telling their employees to use more AI, like their jobs depend on it and in the other hand an educational game dev program is punishing people for using it. I’m not pro-AI, but you’ve got to admit there are some mixed messages going on.
If you’re going to be a developer you will have to learn how to use AI. It’s as simple as that. Obviously there’s a big difference in AI assisted coding and asking Claude to make you a Mario clone.
Stupid in, stupid out applies to teaching as well. 🥲
Better they don’t learn anything at all then the lazy ones pass, right?!
I think this is stupid. Ai use is a symptom and can be controlled. But it is helpful when learning to see more advanced work so you learn how it all comes together.
Edit: I think you’re missing the point: the kids who want to make a game and learn are punished by removing this, and the kids who want to slack off aren’t being hurt by cutting corners.
Later on: “I can’t code or figure out code… how did this happen? I took courses and I chose the AI that passed them… why don’t I know the skills they were supposed to teach? “
No reason to look at all this and assume anyone learns anything.
I work in software. Have for 29 years. There will be no one writing software very soon. Like it or not, seems the ship has sailed. The company I work for has a goal of 100% AI generated code within the next 18 months. So might as well give these kids an A, I guess. It’s not going back to the old way.
People were similarly pig-headed when the first compilers came out. History repeats itself but it’s not really surprising that antis don’t know that or care.
Another software engineer of 25 years here. I’m really surprised to see how badly you’ve been downvoted here for simply telling the truth. I suspect anyone who downvoted you either hasn’t used a paid frontier AI model to write code (so they think the free browser-based models are as good as it gets), assumes you’re pro-AI, or simply jumps on the AI-bad-downvote train.
For anyone that downvoted, tell us, why? Let’s have a conversation about it. Us software engineers are on the bleeding edge of this seismic shift. We’re seeing firsthand what it’s capable of, and how fast it’s improving. It doesn’t mean we’re in favour of the AI companies, or ram shortages, or data centers being rolled out in your backyard, or being turned into engineering team leads overnight. It’s been thrust into our lives just like yours, but if we don’t use it we’re throwing away a massive productivity boost and one of the most transformative technologies of the last few decades.
It’s a rare opportunity someone actually asks for a conversation, so let’s have one.
-
First of all, “no one writing software” taken literally, is simply not true. People still write in all sorts of obsolete / obscure programming languages, mostly for fun, though occasionally there’s reasons why one might want to in a real world scenario. To me, code is art. You cannot stop art.
-
If we’re talking figuratively, when it comes to the “professional” software development. I’m not familiar with this side of things, but here’s my opinion: A lot of the industry has been churning out slop long before AI became a thing. Keeping the gears turning. Making changes nobody’s asked for. Buying up and destroying smaller companies. All in the name of profit and infinite growth. This is bad, AI or not.
-
From what I’ve seen, you at the very least still need experienced programmers to review and guide the AI, otherwise you’ll quickly run into unmaintainable, possibly insecure code. But proliferating AI coding will lead to nobody building that experience.
-
“Insecure code? But just let AI write test cases and review it for security flaws!” It might be good at catching some errors humans make when writing code, but if you spin this to its extreme, you will end up with AI cheating to satisfy the security tests, rather than actually writing secure code.
-
And you say productivity, but all I see is waste. Many people on your side of the fence desperately want to avoid having to label your “work” as being made with AI, because it’s not a good look. People don’t want slopware. AI is good at creating things quickly. To make a lot of noise, drowning out the signal. But in the end it’s just something flashy without substance or longevity.
-
AI is great if you want people further driven away from their fellow humans and into content addiction, with a short attention span, that can barely think for themselves. In the end, it’s the perfect tool for fascism.
-
So, ask yourself: Who actually benefits from this?
I’m not rejecting all that (generative) AI can do. There’s certainly some things that appears to be useful as a tool, in the right hands, for the right context. But the post you’re defending is pushing things to an extreme. And would you know, AI companies themselves are just as extreme. It’s a figurative suicide cult, if not a literal one.
I could go on. And it’s a bit all over the place. But anyway, this is why you’re experiencing a lot of pushback in some online spaces.
-
I see this sentiment constantly and I’ve yet to actually see a single example of AI generated codebase that I’d consider on par with anything but a mediocre one written by humans. You want me to appraise the quality of a product? Show me something that’s high quality, 'cause all I can see right now is billions being poured into products racing from incoherence towards mediocrity.
I don’t agree with this standpoint but I have a genuine question for you and the others in this thread, assuming you’re right: what should we be teaching kids, instead of coding? If the norm becomes not writing code by hand, what would the degree look like that gets you into one of these AI jockey jobs? Do we still teach DSA? Design patterns? Does Programming: AT&T to
3 == "3"get a one-lecture spotlight between punchcard machines and transformer architectures?Or in the long run are you expecting that the entire software industry will essentially become a nontechnical one, and just be middle managers building things directly?
Systems, design patterns, technical experience, and just what good looks like. You need to know how the sausage is made, because you can’t get the best results from an AI model without knowing when the result is good.
And that’s actually really hard without experience, community, and ongoing criticism and learning.
Hey thanks for replying! It’s a great question, and one I don’t have a good answer for.
My immediate gut instinct is that we SHOULD still be teaching kids coding, and everything else that was on their curriculum previously. Just as calculators helped us work faster, we still want to understand what they’re doing. Even if AI can augment or replace some of our knowledge in future, it’s still important to think for yourself and be able to vet what you’re generating with AI. But this is an answer built around what’s happening right now, not tomorrow.
When I think about an answer in relation to my experience over the last year, it feels a lot more hopeless, at least to me. I don’t think we’re going to know what adults should be focusing on in the short term, let alone what kids are learning for their future careers.
The situation is changing so rapidly in my workplace (at least for software engineers) that AI workflows we were using 3 months ago feel ancient. By not being on the latest frontier model, we’re losing productivity. We’re all rolling our own custom software to speed up our personal workflows. It’s faster and cheaper to do this in many cases than to buy an off-the-shelf product and learn/master it.
So already, I feel like the days of companies selling software subscriptions feels very numbered. Adobe and Atlassian are going to have their business model completely destroyed IMHO. But that’s not even the scary part.
Have a look at the number of Linux CVEs found this year compared to previous years - it’s humbling. Governments are paying very close attention, and export controls on AI models are in full swing. They’re very capable cyberweapons.
So the future I’m seeing is an incredibly uncertain one for everyone due to the exponential nature of it all, both from a geopolitics standpoint as well as a career standpoint for many industries. Last year I thought we were somewhere through the hype cycle and it was starting to slow down, but now I’ve had more hands on experience with the frontier models, these days I’m thinking we’re still at the beginning of an exponential curve, an explosion of change that’s going to happen so fast many of us will struggle to cope, and our slow legal and economic systems will struggle to keep up.
I think a lot of people are just assuming it’ll all blow over once the bubble pops, but like all the other bubbles, it’s formed because there is a very real technological leap happening, and it’s not going away.
So right now I’ve never been more sure of one thing: that it’s harder than ever to see what’s coming. So to try and answer your question just feels impossible, it’s all still moving too fast.
Perhaps on reflection one thing we should be teaching our kids is the importance of continued learning and bettering ones self. I fear the world is in for a rough decade from all this (not to mention climate change and late-stage capitalism), and anything we can do to prepare our kids to roll with the punches in that future would be wise.
Thanks for such a detailed response. I suppose asking you to essentially predict the future of this particular industry was a bit ambitious.
My immediate gut instinct is that we SHOULD still be teaching kids coding
As is mine, which for me I think, even if I take it as read that LLMs are the future of programming, comes from the same place that makes me think in spite of Python, programming courses should start with C. But it’s become pretty obvious to me that a lot of people don’t have that desire to “go under the hood”, and still seem to manage, so part of me thinks maybe the same will become true of LLM users.
these days I’m thinking we’re still at the beginning of an exponential curve
a lot of people are just assuming it’ll all blow over once the bubble pops, but like all the other bubbles, it’s formed because there is a very real technological leap happening, and it’s not going away
I’m one of the people you called out who “hasn’t used a paid frontier AI model to write code” (though I didn’t downvote), so I have to take your word for it re the rate of change, but it does seem a little contradictory to me to believe we’re on an exponential slope but also in a bubble. I guess the former is from a technological perspective and the latter financial?
Have a look at the number of Linux CVEs found this year compared to previous years - it’s humbling.
That’s one of the things that has pushed me away from the intensely skeptical end of the spectrum. But I don’t really see how this can fail to flatten out eventually. While secure software is possible at all (by which I suppose I mean cryptography, hardware, and airgaps still work as advertised), I think if anything AI will simply accelerate us towards an equilibrium, where the supply of vulnerabilities that can be found by LLM-guided brute force search is patched and exhausted.
I fear the world is in for a rough decade from all this … and anything we can do to prepare our kids to roll with the punches in that future would be wise.
Pretty bleak. But I definitely agree.
Regarding the frontier models, there’s a idea called Karpathy’s Funnel which notes the massive difference in experience with AI models. It funnels the population of the world like this: (borrowing a lot from a Karpathy tweet with some paraphrasing)
- 6B people (~75% of the population) have barely come in contact with LLMs.
- 1-2B (~20%) are casual and infrequent users of free-tier ChatGPT-like products. This group sees derpy chatbots and AI slop, to this part of the world the whole thing feels crazy still.
- 20M people (0.2%) see frontier-grade LLMs, complex projects that used to take them weeks/months can now be completed by agents with a prompt. You typically need a budget to get this experience. There are multiple levels in this tier as well from those that use ‘open-weight’ models to expensive monsters like Claude Fable. Models are Powerful tools at this level.
- 5k People (~0.00006%) have access to frontier-grade systems internally inside AI labs, they have the budget to put the frontier on automatic, spin up hundreds of agents, and burn their huge salary’s worth in tokens.
At these lower tiers of the funnel, they believe in the future the AI companies promise because they’ve literally bought in enough to actually get the experience.
How soon are we talking? Tons of people reject AI, and they continue making software.
I’m sure even in companies things are going to change, as it’s already not cheaper to make software with AI, not to mention the technical debt they’re building with nobody understanding the code largely produced by AI. Sure, as much as I want it, AI is probably not going to go away completely, but it’s going to swing back hard in the other direction.
not to mention the technical debt they’re building with nobody understanding the code largely produced by AI.
Swap AI with humans and this statement continues to be true.
