At the chalk face: Will AI replace coding in schools? Why coding is here to stay..

Will AI replace coding in schools? Or is programming more important than ever?

In this episode of At the Chalkface, Craig and Dave dive headfirst into one of the biggest debates in computer science education right now: with AI and “vibe coding” exploding in popularity, do we still need to teach students how to code?

From the latest curriculum and assessment review to the growing influence of generative AI tools, it’s easy to see why some are questioning the future of programming in the classroom. If AI can generate working code in seconds, why are students still wrestling with loops, conditions and syntax errors?

Craig and Dave unpack why that thinking misses the point entirely.

This conversation goes far beyond writing code. It’s about what students actually learn through programming: resilience, problem solving, creativity, and the ability to debug messy, unpredictable outcomes. Whether it’s tackling a stubborn bug or building a solution from scratch, coding remains one of the most powerful ways to develop computational thinking skills that AI simply can’t replace.

They also explore the risks of over-relying on AI in education and industry, including the loss of junior developer pathways and the deeper issue of accountability when AI-generated systems fail.

Plus, Dave introduces a bold idea for the future of assessment: Code AI – a new framework for rethinking NEA that embraces AI rather than fighting it.

If you teach computer science, work in education, or just care about the future of tech skills, this one is a must-watch.

Don't forget to like, subscribe, and turn on notifications for more conversations from At the Chalk Face.

🔗 More from us:
Website & resources – Craigndave.org
Smart Revise – smartrevise.online


#AtTheChalkface #ArtificialIntelligence  #DigitalLiteracy #EdTech #Teaching #ComputingEducation #ComputerScience  #AIinEducation #Programming #CraigAndDave #EdTech #VibeCoding #TeachingComputing  #coding #codingforbeginners #computercoding 

Transcript

SPEAKER_00 0:00

Will AI mean we don't have to teach coding anymore?

SPEAKER_01 0:07

Hello and welcome back to another At the Chalk Face with myself, Craig, and my colleague Dave. If you're new to us, we are both ex-teachers of computer science in the county of Gloucestershire, and we run the company Craig and Dave, providing a one-stop shop solution to everything you need for computer science at secondary schools. And if you are familiar with us and you've come back, then thank you. Obviously, we haven't completely bored you. So we're gonna hit you with um a very interesting topic today, and we're calling this video Coding is Here to Stay, because boy are we hearing a lot of there's no point in coding anymore. Yeah, AI means we don't need programmers, and our own lead developer has quite a lot to say on this. We should get him on one day, Dave. But um, unless you've been really living under a rock, obviously the massive car report, the curriculum review, which of course we start teaching the key stage three from 2028 and then key stage four and five rolling on after that, made it super evident that programming is here to stay. Uh, it's a fundamental thing, teaching algorithms. And we're going to unpick that a little bit because you know, are there things AI can do for you? Of course there are, but there's nuance in here. And I think uh what Dave and I will argue is there's a lot more skills uh around programming that students pick up beyond the actual programming, and by handing it all off to an LMM, you're losing a lot of extra stuff, and that's what sometimes gets lost. So let's get into that a little bit, Dave. Do you want to start with an opening statement or gambit of your own, or do we dive straight in?

SPEAKER_00 1:45

Well, I think that perhaps it was a little bit surprising when we saw the curriculum and assessment review, and then we saw the government response to that. And we know that the uh British Computer Society are drafting the new uh National Curriculum for Computing and programs of study. And I think there were plenty of teachers out there at the time that thought, oh goody, this is an opportunity to potentially drop programming as something that's in the curriculum. Because, hey, why do students need to learn to program anymore when vibe coding is such a big thing? And yeah, you know, let's face it, vibe coding is a massive thing at the moment, and we've seen it ourselves. Uh, it used to be that creating revision tools like Smart Revise was, you know, the gift of the professional programmer. And, you know, we invest in a lot of development capacity in order to create very robust and secure and useful tools for students. But we're seeing these little micro apps popping up all over the place that um are doing similar sorts of things, and so uh and they're being written by people with you know, sometimes very little coding experience and uh and putting these things uh together very quickly. And so people I think were understandably saying, well, do we do we really even need to teach programming anymore? You know, um, and I know there's a lot of uh teachers out there that find programming perhaps one of the most challenging aspects of computing, whether that's at key stage three or GCSE, and certainly at A level, we've got a lot of non-specialists um in our subject. And we've done episodes on non-specialists before where we've said, you know what, are they a cure, are they a curse? And you know, I think we concluded there that uh actually non-specialists um have upskilled quite a lot and have brought quite a lot to our subject. So we're certainly not uh doing down our non-specialists at all, but we do know they find programming really hard, and I'm sure they would jump for joy if they thought that they didn't have to teach uh programming, and similarly for for students too, I reckon, Craig.

SPEAKER_01 3:59

Yeah, but that that's the crux of it right there, though, Dave. I think you've just hit on a lovely springboard into the rest of this episode. I think the, and you know, no disrespect to the teachers of finding challenging as right. We have an awful lot of non-specialists, people that have converted. It is challenging. But I think the teachers and students out there that wish it was dropped because of the challenge are missing the point of the benefits we will lose if we hand all the programming over to AI. So I'm gonna kind of pose a couple of statements or questions here, which beautifully link us on to where we were going, because you know, coding itself, you know, the act of sitting at the keyboard and you know, it's a lot more than simply entering a bunch of commands. It is all those skills we want to embed in students. It's all that cognitive training, it all comes in programming, not you know, giving up resilience, problem-solving, creative thinking. And when students hit a bug, is a bug simply an error? No, a bug is a puzzle to tackle. Again, applying so many of these skills and resilience and independence and you know, adapting and reasoning, it it's it's there's so much more to it. So I think you provided a lovely opportunity for us to springboard off. So, yeah, I mean, some of that stuff then. I mean, I I pose kind of two statements there, you know, coding simply isn't the act of entering commands, you know, it's it's a lot more than that. And uh, you know, when you hit a bug, it's not just something to put your hand up and say, Miss, come and do this, it's a puzzle to be solved. I think this is the essence of why programming is still so important.

SPEAKER_00 5:30

Yeah, I uh absolutely, absolutely, and I think um that gets lost in translation uh too frequently. I think um students don't really understand that there is so much more to programming than, as you say, just simply entering commands. And and for teachers too, at a very shallow superficial level, you could look at programming and say, you know, we we don't really need to teach the basics of um conditions anymore, you know, if statements and you know, loops and repetitions, you know, while statements, for loops, that kind of thing. You know, why why teach the absolute fundamentals of programming when let's face it, there isn't a single um GCSE or A-level problem that AI can't solve in a matter of seconds? It it can, you know. I think we've said this before, but thinking about our time to code uh programming uh pedagogy and set of exercises, AI can solve every one of those exercises in under a second. Yeah. Um so you say, well, why bother when you know AI can just do all the basics for students anyway? And it it really misses the point. Yeah. Um, and I find it a bit frustrating actually when students and teachers say, I don't see the point of this, or or you know, they they email us and say, Oh, I managed to solve your time to co problems by just putting it into AI. Yeah, well then, you've you've sort of missed the point. We're not creating programs that are impossible for AI to solve. Uh, what we're doing is creating programs that cause you to take that problem-solving approach that you just alluded to, that idea of having a set of requirements and thinking about the steps that are needed in order to provide a solution to meet those set of requirements, to have a sort of a hypothesis of um something that you think might work, you give it a go, and inevitably it doesn't work for all sorts of reasons, whether that's a simple syntax error or something more fundamental that you didn't see initially, but as you've gone down a particular path with an approach, it you know, it starts to get worse and worse, and the problem sort of unravels a little bit, and you're like, Oh, I can't go any deeper with this. I need to chuck all that work away and go right back to the the start sometimes and and take a different approach. That that resilience that you need, you know. I mean, we've all seen it, Craig, right? In coding lessons. Um, you know, the hands go up, uh, you know, so I've got a syntax error, so I can't say, so I've got a problem, so I've got certain it's really, really frustrating, right? And so I think programming is that opportunity to teach the students the resilience. It's like you're not going to get the right answer first time. You've got to keep going and going and going. And you mentioned creativity as well, and just you know, this idea that there are hundreds of ways of solving a problem, uh, particularly with programming. And again, it's it's a source of frustration for teachers, particularly again, our non-specialists, who want to know the answer. And when you say, Well, this isn't maths, there isn't one answer. There are lots of ways you might have solved this problem. Some are better than others, you know, some might be more secure, some might be more robust, some might use memory more efficiently, some might use the processor more efficiently. There are so many different ways of getting to the answer, and that's about the creativity within the subject.

SPEAKER_01 9:12

Yeah, and we've you know thrown a lot of key words out there, but one really stands out more than the rest for me. And you know, you could make a case for any of these over the others, I guess, you know, but we've thrown around problem solving, being analytical, creative, adaptation, all these cool things. But the one that sticks in my head more than anything else, the one I hear more and more teachers complaining about, and the thing I feel we're you losing in our young people as the use of social media and instant, you know, gratification via little tiny short TikTok memes and everything else, is that resilience. For me, that's the one that really sticks out. I feel that that young people are losing the ability to be resilient, and that is something you will enforce in the challenge of programming, probably more than any other area of the computer science or just computing in general specification. Yeah, I mean, yes, there are there are tricky theory topics and you can have fun debates and stuff, but I think that that real ability to be like, right, I've it's not working, I'm stuck. This is a problem for me to solve, a problem for me to persist with. What's the first step? What tools do I have in the toolbox to help me overcome this and not put my hand up and wait possibly five, seven minutes before the teacher gets to me? I think, you know, I think don't think for me, Dave, there's any other area of computer science that can build as much resilience as programming. You know, and the other beauty, and this isn't isolated, obviously, to computer science as a subject at all, but I mean, um, with all subjects, but especially programming, um, it allows students to make mistakes and realize that that's part of the learning process. So, you know, if programming's taught well, it gives them an environment where making a mistake, you know, having an error, failing is an acceptable part of learning. Okay. So I learn when I well when I fail. Where do I go from here? How do I overcome that? So for me, those those things are probably one of the biggest reasons why, you know, I think at least programming from an educational point of view needs to stay. And it is staying. I mean, it's as clear as clear as day.

SPEAKER_00 11:21

So and you know, there's no denying it that students do find it the most challenging part of the computing course. And I'm sure some students are definitely put off with computing because it's got um a large programming element, and because we know, Greg and I, that uh students struggle so much, we make it an integral part of our GCSE and A level right from the very first lesson. So, although it might only be a small part of the specification, and that changes from course to course, yeah. Um, but it's you know a smaller part of the specification in in many courses, particularly at GCSE, uh with OCR and well, OCR mostly, um there's a temptation there to not spend a lot of time on it. And of course, you and I spend you know 50% of our lessons on programming, even though it's not 50% of the specifications, and because it's just one of those skills you have to practice with. And as you were talking, I I was kind of building um this analogy in my head and thinking, do you know what? It's a little bit like jigsaw puzzles, isn't it? You know, you can imagine giving a jigsaw puzzle to a student saying, here, here's a box of a thousand pieces, uh, you know, create the picture. And they don't really struggle with that. I mean, what I mean by that is yes, they don't instantly get to the solution, but they're quite happy to say, Oh, well, let's start with the corners because that's the the easy bit. So they go hunting for the corner pieces and they put those in. Oh, let's now go do and do the edge pieces and they put the edge pieces in, and then they slowly build up the middle by building one piece on top of another, you know, not just putting some random piece in the middle and hoping for the best, but building the picture over time by locking one new piece into an old piece. And I haven't really thought about programming like this until I was listening to a moment ago and thinking, do you know what? This is a really good analogy, perhaps even to share with students and say, look, programming is a jigsaw puzzle. Yeah, you've got the solution you want, you've got all these pieces, but we've got to know sort of where to start and which pieces lock into which pieces and build the picture over time. And crucially, you alluded to it a moment ago, you're going to make mistakes. Yeah, you're going to pick up a piece, you're going to try it, and it doesn't quite fit. So you're like, oh, okay, that one doesn't fit. And you know, that's making mistakes and then trying again and persevering. And as I said a moment ago, you're not going to get that jigsaw puzzle solution instantly. It's going to take a long period of time. And I think encouraging students into that mindset. And what's really interesting for me though, Craig, is I guarantee you, if I was to give jigsaw puzzles to students, and I'm increasingly thinking maybe I would do this as an introductory exercise to programming, although it would be a nightmare collecting all the pieces back in again. But if you did that, um no student would struggle with it. Yeah now they might get bored of it. I'm not saying that, but they wouldn't struggle with it. And my question is, why do they not struggle with a jigsaw puzzle, which has so many similarities with programming, but they do struggle with programming. I'll tell you what I think it is. It's purely familiarity. They've been doing jigsaw puzzles since they were before primary school, right? Just building puzzles with like four pieces, six pieces, ten pieces, and then working up to 50 pieces, right? And as they've got older, those programs and those jigsaws have got more sophisticated. But the point is, they've been doing it from such a young age that by the time they're teenagers, yes, they might get bored of it, but they don't see it as a challenge. Yeah, and I think there's something in that.

SPEAKER_01 15:00

Yeah, absolutely. My god, I I could run away with the analogy, but but I but I won't, you know, you're going through the box of a hundred pieces and like, oh, there's a corner piece. I know I've only got four, it must be one of these. I'll do that. These are edge pieces, and of course they start building the structure. I'm now thinking of like the layout of the programme. It's like, you know, you you're right, but those things are inbuilt. That's due to repetition. And then you obviously the resilience, it's it's yeah, very interesting. Why haven't we done that as a starter to programming before? You're you're right, though, I'd have puzzle pieces everywhere. But yeah, absolutely. And it all comes down to this resilience for me. Um, and of course, there's a there's a there's a bigger problem to all this. I mean, you know, we're turning around and saying all these valid reasons why programming is useful in school, beyond just purely, you know, the act of programming, all the problem solving, computational thinking, you know, learning to make mistakes, resilience, all of that. But there's a bigger problem. Let's say we accept the fact that, okay, it's it's there, and let's say it wasn't in the curriculum and therefore we all drop it. And uh, you know, what actually hands up what actually ends up happening in industry to this, you know, this this pipeline of new programmers coming in, and old ones coming out, and it doesn't take long to search on the internet before you find there is already plenty of evidence out there that the big companies who rush to lay off a load of their developers to save money as AI are now quietly hiring them all back and quite often at a much higher fee. That kind of AI scandal there of letting programmers go is already starting to burst, and the companies are realizing it wasn't quite where it was. So, you know, you don't hear about that, that doesn't really hit the news, but if you search for it, you will find it. Um, and we can't perpetuate that problem. So, you know, what what what's going on there, Dave? Because obviously industry jumped on this and went, oh, we can fire our programmers, they're paid an awful lot of money, and AI can do most of it. Why are they now hiring them all back? What's the problem?

SPEAKER_00 16:57

Yeah, well, it's symptomatic of that short-term profit-driven thinking where it's like, oh, you know, we can make more money if we uh pay less staff, and actually we can increase our productivity too, because what a machine can do in the same time as a human, and the human isn't gonna get the human's gonna go off sick, go off on holiday, all the rest of it, and we can get these we can get these machines working 24-7 until they break down. But you're absolutely right. What it's created is um a problem with the pipeline, as you put it, because you had your you know, school lever programmers, let's call them that, that go into university, become graduate programmers, they come out into the industry as junior programmers, and then through lots of experience and exposure to different problems, they become mid-level programmers, and then eventually, you know, they become senior programmers. And those senior programmers, it's a real skill set, it's a different skill set. It's not just about programming once you get to that senior level, it's that deep analytical thought and knowing what the pitfalls might be because you've seen them and been exposed to them for so many times over the years that it would be unrealistic to expect a junior programmer to spot simply through a lack of experience. But you could argue, yeah, but AI, you know, can soak up in its training data every single program that's ever been written, and you know, it it can it can accelerate that process um really, really quickly. And there's no technical reason why ultimately AI couldn't become a senior developer. And I I think that might be true. I know there's a lot of skeptics out there that say, oh no, you can never replace the true human senior programmer. I'm not sure about that. I think technologically that might be possible. Um but I think the the big issue for me is there's who's accountable when it comes to editing? So I'm thinking about vibe coding again. You know, people put together these solutions very quickly, they put them up onto the web, they expose them to the to the world. But who's taking responsibility for the security? Who's taking responsibility for the data protection questions? You know, who where where's the where's the ethics in this? Who who ultimately is responsible and accountable when these things inevitably get exposed one way or the other? And and it will happen. I think there's a naivety out there, particularly with vibe coding at the moment, that oh, it you know, it's fine, it's robust, and you know, AI is better than any program right there, it's it's not going to be a problem. Well, there's always problems, it's just how long does it take to expose the weakness?

SPEAKER_01 19:52

Yeah. And then when you do find that weakness, and you need a senior programmer to intervene and have a look at it and fix it, if we've got less and less of them, I mean there's a there's a problem. You know, these senior coders won't be senior coders forever, they'll retire, they'll they'll move on. And if we don't have ones replacing it, because when you were hitting on it earlier about the the lifetime of experience with senior programme, what it means, you were making me think of um uh Mark, who we've had on a couple of these episodes before, our own uh senior developer here at Craig and Dave. And I mean, Dave and I have been programmers our entire life. We are hobbyist programmers as kids. We went to university computer science degrees, we teach programming, but when I tell you our analytical skills are not even a touch on Mark's, I am not underestimating. And it it drives us, sorry, Mark, it drives Dave and I so mad sometimes, but almost in awe. We will have designed a problem, looked at it, and we're computational thinkers, we're logical problem solvers. We look at the edge cases and think about this, and we present a beautiful design to Mark, and we think, you know, he'll spot some things we've missed, and he'll see a design raw out the door, just talking it for the first time, and uh he'll sit there and go, What about oh, here we go, but you can't deny it. I mean, it is phenomenal, and that has been built over 20, 30, 40 years of working in the public and the private sector programming. I mean, that level of analytical computational thinking has come from years and years and years of programming, and um, you're right, why it might be technically feasible to get the AI, I'm not sure we're quite there yet. It doesn't answer a lot of those questions. How do you hold the AI to account? You know, as you say, the ethics, the moral issues. So, no, and do we just want to lose that level of resilience in computational thinking and analytical problem solving from our young people anyway? I mean, I would argue no.

SPEAKER_00 21:47

And at the moment, uh I I I've got a slight fear um of I don't know how to describe it really, the fear of the black box, right? At the moment, yeah, um, any experienced process. Programmer can take the hood off a program and they can work it out, right? I'm one thing that I'm particularly intrigued at watching videos on YouTube about people sort of reverse engineering really old games and working out kind of how they work, and as a result of doing that, then re-engineering them back to a new platform, right? Um, I find that just just intriguing, but it it illustrates the point that you can take the hood off any program and you can work out um exactly what's going on. And my fear, I suppose, and and I don't know if it's a true fear or not, because ultimately everything comes down to ones and zeros. So is it not possible to re-engineer everything? I I don't know the answer really. But this idea that when AI is creating things, um, it's not possible to take the lid off and explore and fully understand what's going on. And sure, it is at the moment, because you could just, you know, get the AI to produce Python, produce C, whatever it is. I get that. Yeah, that's where we are at the moment. But it could well be that in the future, um, that is a blocker to true progression. The fact that we are constrained by a language that was written by humans for humans to make it understandable could be a limiting factor for machines. And in and I don't see why they wouldn't then create their own language, for want of a better term, that humans couldn't understand in the aim of efficiency. So I don't know. I mean, we're getting into sort of like deeper expressions.

SPEAKER_01 23:43

I I won't go off, yeah, I won't go off on a long one here, but it's interesting you say that. I don't know if you've you've seen this come along. There was an interesting experiment that Google did, and I read all about it, where um they got two AIs to talk to each other, and within about sort of 15 seconds or 20 seconds of chat, they both realized they were talking to another LM. So they developed a much faster way of communicating in the chat. Have you seen that? Yeah, that was really interesting.

SPEAKER_00 24:08

It reminds me of teenagers doing the similar thing with SMS back in the day, right?

SPEAKER_01 24:14

Yeah, yeah. That that's it. I just thought it was fascinating. It was like really quickly went, Oh, you're an LM, you're an LM. Oh, we don't need to write and respond in this verbose English nature. No, let's be more efficient and carry on the dialogue in a way that's more efficient for the pair of us. And they naturally spoke in a much quicker form, which then you couldn't understand. Uh really interesting. I worth looking at that. That was fascinating. I do have an interesting question for you though, Dave. Yeah. So let's let's get down to like, you know, because we've talked a lot in the abstract. We know that um programming in terms of an assessment tool will probably most likely remain out of the new GCSE of computing, but will stay in the A level as a component. So assuming the A-level NEA is still going to exist when the new ones come in and there's a programming project, an independent project of some form, then what on earth does that look like in today's day and age with AI? Because a lot of this is done outside the classroom in in A-level students' independent time. You've got to accept, I mean, AI is already here, and by the time the new A levels roll in in four years, it will be you know even more infused. So, what do we do about that? Do we just not give any marks at all for AI code? Uh, does the project go? I mean, you always have thoughts on these sort of things, but how do you see the the NEA evolving to incorporate these, you know, these issues with AI?

SPEAKER_00 25:39

Yeah. Well, I think firstly, it's a big assumption that the NEA will stay. I think the actual hunch is it will go and it will be replaced with an exam. Because I think I think the thinking around this NEA.

SPEAKER_01 25:53

We don't know that yet, though. There's nothing out there saying that yet.

SPEAKER_00 25:56

No, no, no, we don't. This is just a hunch, right? I I think the NEA will disappear. Um, and I think that's because the thinking around the NEA is wrong. I think at the moment, uh, the big problems that are being identified are that A, the NEA is taking the students too long, the length and the size of the reports and the write-ups that they are making has got wildly out of hand. And there's this fear of AI that oh, you can only credit the things that the student has done, not the things that they've done with AI. And that and that's where we are at the moment. And I just think that's um so such basic thinking. These problems are not difficult to solve, and I just think that it would be a real shame if we lose that real creative aspect in the assessment of A-level computer science, yeah. Because for me, it's it's part of the fun of the subject, and if you just replace that with just pure theoretical exams, I think we've really lost something. And I know there are lots of teachers out there that will be jumping up and down looking forward to the day when there's no NEA. I know that, but you know what why are you hoping for that? I think you're hoping for that because A, you hate marking it, it's a massive burden, right? And and B, you're you're frightened of your students cheating and you don't know uh where the accountability lies and it it can create problems, and you probably battle with some students trying to even get them through this NEA, right? I get I get it, but I don't think those problems are insurmountable. And what I would like to see is instead of us making AI the problem, let's make AI the solution, and let's actually embed AI right at the heart of the NEA. So what we do is we say to the students, look, this is a new way of thinking about the NEA. What you do is, and I've got a little acronym for this, Craig.

SPEAKER_01 28:06

Oh, of course you have an acronym. You love your acronyms. Okay, go ahead.

SPEAKER_00 28:10

And it's called CODAI, right? Very clear. And so it's C O D E A I, right? And it stands for create, orchestrate, debug, experiment, adapt, improve, right? So you heard it here first. This is how I would like the exam boards and off qual to move the NEA forward. So what we do is we create a solution to a problem, that's the create, by working with others and AI. So orchestrate, put AI at the heart of this, okay. Debug because whenever you get AI to create something through vibe coding, it never quite does what you want straight out of the tin. And so you still have a debugging process, but it might not be at a syntax level. Yeah, it might be at a syntax level for the for the AI, where the program doesn't work properly or it doesn't run properly, and you say, Ah, actually, you know, um it does this, but I wanted it to do that. I see that as a higher level debugging skill. We don't need to debug individual syntax errors anymore. Yeah, and then experiment, that's my E. So by finding better solutions, so you solve part of the problem, and then you think, ah, but it would be even better if it did this and it prevented these kinds of inputs from happening, which you know just don't make the program work the way I want it to. So you're experimenting and trying lots of different ideas because vibe coding can be done so quickly. Let's harness that power in creating quick solutions to experiment with different ideas and then adapt A. So evolving the code generated, um, as I say, into extending the solution further, which is then my I improved to make the solution better. So, really, what I'm describing is a fully iterative vibe coding approach, where what you're really assessing is the student's ability to understand the specification and the requirements, to produce a solution by using AI that fulfills those requirements, but then going beyond that, and this is where the higher grades would come from. Going beyond that to understand where the solution may need to be more robust, where it lacks a bit of security, where you can improve the solution for the user in ways that perhaps they hadn't originally anticipated. That's where you get your your top marks. And just to conclude this, Craig, I at the moment we're kind of constrained by this almighty write-up. Well, the solution is important. You just say to centers and to teachers and students that you create a development diary that could either be a video or it could be hand, it could be typed, you know, well, whatever you like, it doesn't really matter. But crucially, you say it can only be six pages, and each of those six pages, no more than a side of A4 each, must demonstrate the six skills of create, orchestrate, debug, experiment, adapt, improve. So you only ever mark six uh pages, no more, and then you specify something like it must be 11 point, right? So you don't get these crazy situations, and then with the video, you just say, look, it's a maximum of six minutes because it's one minute per area in which you discuss, and then it's down to the finesse of the student to choose which parts of what they've done best exemplify those six areas in order to uh fulfill the requirements. And so for me, this is a beautifully elegant solution to the NEA. I've fixed the problem of it getting out of hand, I've fixed the problem of it not having the structure and the scaffolding um that's required, and I fixed the problem of AI because AI is fundamental to what you do.

SPEAKER_01 32:21

There you are, Dave. Solved the future of the NEA. You're welcome off quo, JCQ, and all the exam boards. You heard it first, code AI. Well, we'll see. I mean, let's be fair to the exam boards. We we have a close, you know, close relationship with all the um subject advisors from the major exam boards, and and we know at least one or two of them that are thinking along these sort of lines that we can't just, you know, that the horse is bolted. We could we, you know, with the evolution. I mean, obviously they're all waiting on guidance uh from higher on because they will have to work within the constraints of the DFE. But you know, if they have their way, they're saying, no, you know, the next evolution will need to embrace AI in a sensible way. We can't simply ignore it and say it must all be done under control conditions with the internet turned off, and oh dear, we tried that back with the NEA at ECSE and what a disaster. So uh, but we will see, won't we? We will see. They will be uh confined by the constraints of the DFE. So let's hope that sensible people are saying sensible things to sensible people. Dave's like skeptical. Uh any closing thoughts then? Interesting discussion. Any any closing thoughts before we wrap up, Dave?

SPEAKER_00 33:27

Yeah, so my question really is now to the audience. Put your comments below. Do you agree with us? Do you think programming is something that we should teach? And if so, why? Do you think we should scrap it? And if so, why? What do you think about my vision for the A-level NEA? How would you like to see that changed? Or would you just like to see it scrapped completely?

SPEAKER_01 33:48

Brilliant. Well look, thanks for joining us. As always, we'll be back next week with another video. Take care for now. Bye bye.