Education Committee
Tuesday 8th September 2026
Skip to latest contribution
This debate has concluded
(began 6 days, 5 hours ago)
Watch LiveSkip to latest contribution
This debate has concluded
10:02
Helen Hayes
-
-
Helen Hayes
Order. Welcome to this morning's evidence session of the Education Select Committee. This morning, we are holding our second oral evidence session in our inquiry on AI and EdTech and their role in our education system. I'll invite our witnesses for this morning to introduce yourselves to us. If I should say at the outset, we've got three panels of witnesses this morning, which is an unusually congested agenda for the committee. So I'll be reasonably strict on time, and I would appreciate if witnesses could, and indeed, members in your questioning could keep your contributions as succinct as possible, because that will help us to get through all of the topics that we need to.
But, Professor Clark Wilson, if I can start with you. Thank you.
But, Professor Clark Wilson, if I can start with you. Thank you.
10:02
10:03
10:03
10:03
10:03
10:03
10:03
10:03
10:03
Speaker 2
-
Speaker 2
I think it's having an increasingly significant impact, both in terms of the AI that children are accessing at home, and also the kind of increasing prevalence of AI in the education system. I think we might explore this a little bit in more detail later on, but I think there are definitely advantages and positives to that. But there are also risks that need to be managed as well. But I think clearly AI is becoming a much more important part of children and young people's lives, and also an important part of their education.
10:04
10:04
Speaker 3
-
Speaker 3
Yes, so the Commission for Countering Online Conspiracies in School, published first in 2024, to report of the growing problems of misinformation and conspiracy theories in schools. And they recently did a survey and published new findings, which showed that there has been a significant growth in the problems conspiracy theories are presenting in schools, from pupils to teachers, and that growth is largely been identified as being driven by AI. So I think that's a particular aspect that's often overlooked when we're thinking about the impact of AI in schools and what we need to consider.
The second point, just to add as well, in terms of the teachers that we speak to on a daily basis, we know there are many fantastic teachers who are using the technology in a very judicious and thoughtful way to try and enhance the quality of their planning and their resources. But we know there are a huge number of teachers as well that are ethically deciding to opt out of use of AI because of its concerns environmentally, ethically and cognitively as well.
So we are a very mixed point for the sector in terms of how AI is impacting in schools at the moment.
The second point, just to add as well, in terms of the teachers that we speak to on a daily basis, we know there are many fantastic teachers who are using the technology in a very judicious and thoughtful way to try and enhance the quality of their planning and their resources. But we know there are a huge number of teachers as well that are ethically deciding to opt out of use of AI because of its concerns environmentally, ethically and cognitively as well.
So we are a very mixed point for the sector in terms of how AI is impacting in schools at the moment.
10:05
10:05
Rebecca Paul
-
-
Rebecca Paul
I think it reflects I totally agree with Darren here in that it's impacting , and I think the rate and the scale to which it's impacting very much depends on the approach to innovation in the different institutions, schools and colleges, those who are always at the front and have a, I would say, a more innovation focused with processes to think about how it's impacting are in one place. Whereas we have many places, many institutions which are sort of holding back to see what others are doing.
So quite a mixed bag.
So quite a mixed bag.
10:06
10:06
Speaker 1
-
Speaker 1
Yeah. So I think at the moment, a lot of the technology that we're seeing in schools is very much designed to kind of automate and then theoretically automate, augment teachers work. But at the heart of a lot of these claims is a very strong efficiency logic. And not really thinking about teacher autonomy or teacher expertise. So in some extensive ethnographic work that we have done in secondary schools, we've really seen how these automation logics have compromised teacher expertise.
And this can happen, for example, in the ways that they kind of remove information for teachers about how to teach their subjects. So for example, when they're using AI for lesson planning or remove information about how to teach their subject in terms of the kind of black box nature of the AI that they're using. And we're also sort of seeing kind of contrast to augmentation. We're also seeing teachers spending an awful lot of time making this technology work for them.
So dealing with tech failures, dealing with tech kind of passwords, logins, all these kinds of things. And so this means that it's not really augmenting in any kind of meaningful sense. And it's kind of reducing, you know, the potential for teachers to be autonomous and expert in their teaching. And the last point I'll make very briefly is that I think also there's this claim that technology is going to supplement teachers and enhance what they do.
But what we're seeing, particularly in schools that are more serving more disadvantaged communities and have less resources, they're starting to rely on these technologies to some degree. And there is some forms of automation happening in some schools.
And this can happen, for example, in the ways that they kind of remove information for teachers about how to teach their subjects. So for example, when they're using AI for lesson planning or remove information about how to teach their subject in terms of the kind of black box nature of the AI that they're using. And we're also sort of seeing kind of contrast to augmentation. We're also seeing teachers spending an awful lot of time making this technology work for them.
So dealing with tech failures, dealing with tech kind of passwords, logins, all these kinds of things. And so this means that it's not really augmenting in any kind of meaningful sense. And it's kind of reducing, you know, the potential for teachers to be autonomous and expert in their teaching. And the last point I'll make very briefly is that I think also there's this claim that technology is going to supplement teachers and enhance what they do.
But what we're seeing, particularly in schools that are more serving more disadvantaged communities and have less resources, they're starting to rely on these technologies to some degree. And there is some forms of automation happening in some schools.
10:07
Helen Hayes
-
-
Helen Hayes
Do you think? I'll start with you again, professor? I know, but others may wish to chip in. Do you think there is a need to think about the fundamental purpose and role of teaching in light of the prevalence of this technology, and not, therefore, to equip teachers with traditional teaching skills and hope for the best in the context of AI. But to really think in a more profound way about what teaching means in this, in this context and, and what would that what would that look like if that's the case? Yeah.
10:08
Speaker 1
-
Speaker 1
So I think very much so we need to start to change the ways we're thinking about what teachers, what teachers skills might mean or what teaching looks like in these contexts. And I'd really like to see a much stronger basis for these discussions based on teacher expertise. What we're seeing a lot of is a kind of agenda being set by the commercial companies that build these products, rather than the teachers or the education community themselves. And I think what we're really needing to encourage in our schools and in our education system is a kind of critical literacy for teachers that really allows them to have the language and space and time to really, you know, critique and push back and, and sometimes refuse, but also ask for alternatives or different kinds of technology.
So there's something about a kind of shift from an agenda set by kind of technological kind of commercial agendas towards something that's much more educative. Yeah.
So there's something about a kind of shift from an agenda set by kind of technological kind of commercial agendas towards something that's much more educative. Yeah.
10:09
Rebecca Paul
-
-
Rebecca Paul
Thank you. I'd go beyond the teaching and teaching practices in terms of the changes that are needed, because I think unless we fundamentally look at the curriculum, look at the purpose of education more broadly and really think about what we value and therefore how we assess what we value, I think we're tinkering if we don't really take the bigger piece here, because it's undoubtedly changing everything. Therefore, we have to really think about what we teach and what we assess, because we know that that drives so much of what goes on in schools, colleges.
10:09
10:09
Speaker 2
-
Speaker 2
I think one of the things that our members report to us is the risk that they encounter, that technology leads what happens in classrooms, rather than pedagogy and thinking about what are the aims and objectives that you want children to secure. That's quite a long standing problem. I mean, I can think back to when I was teaching an interactive whiteboards came in and they were installed in every classroom. And then after that we thought about, well, how are we going to make use of this? Rather than thinking about how can technology support learning and achievement, that technology arrives, and then we try and retrofit educational objectives to that.
That's, as I say, that's quite a long standing pattern in the education system in England. But we see that very strongly in respect of AI.
That's, as I say, that's quite a long standing pattern in the education system in England. But we see that very strongly in respect of AI.
10:10
10:10
Speaker 3
-
Speaker 3
just like to add that I think the idea that using AI in education being inevitable, one is one that we need to question. It is a choice that we as a sector are going to have to make. And while there are many schools are choosing to use it, there are many that are choosing not to use it. And we need to look at the reasons why and not assume that we need to adapt what teaching is to accommodate a technology, that we have a choice about whether we want to use or not.
I'm not denying AI is not prevalent and changing society at large, but we do have a choice about whether we want to bring it into our school system.
I'm not denying AI is not prevalent and changing society at large, but we do have a choice about whether we want to bring it into our school system.
10:11
10:11
Rebecca Paul
-
-
Rebecca Paul
I think the answer is undoubtedly yes, because without really understanding what are the working practices, the professional working practices of educators that we might offload to and essentially we're offloading to a technology. I think the risk comes from a lack of trust in the in the school classroom parental environments around who's making judgements, who's providing feedback, who is creating the lessons. And if we don't have that very explicit and discussed and agreed within schools and institutions, I think that's where we can lose the trust.
And that won't help the profession.
And that won't help the profession.
10:11
Speaker 1
-
Speaker 1
Yeah. So I just echo what I've said in my opening response, but I absolutely think that we really need to think about these questions of expertise and autonomy. And definitely in the work that I've done where we support teachers, kind of recognise their expertise and crucially, support teachers autonomy in deciding how to use technologies, when to use technologies and how to teach. That's when we see the most educative uses of technology.
10:12
Speaker 2
-
Speaker 2
I can't agree with those, those two contributions really. I think it is something, again, that teachers are very concerned about, that technology, particularly AI driven technology, replaces that professional judgment and discretion over children's learning, pedagogy, assessment, planning, all of those key professional functions. I think teachers, many teachers are nervous that what's that? What could happen and perhaps in some instances is starting to happen, is that those professional functions are being displaced by AI in a way that not only undermines teacher professionalism, but will undermine the learning experiences of children and young people.
10:12
10:12
Speaker 3
-
Speaker 3
And I think we need more investment for test and learn in this space urgently, and an expansion of the remit of the funding that's been committed for research in this area to include teacher development, which is currently not defined within those parameters. We know there are real problems in the sector that we want to fix. One, for example, is how to ensure all trainee and early career teachers have access to high quality, consistent mentoring really difficult for schools to provide that provision.
So is there an opportunity for AI assisted observations to support and allow expert teachers to see what's happening in novice teachers classrooms and provide support? The simple answer is we don't know what the impact of that will be, but that's a problem that needs to be solved somehow. So why don't we test the technology and evaluate what the impact is? Because there could be great gains, but there could be unintended consequences which cancel out those great gains.
So is there an opportunity for AI assisted observations to support and allow expert teachers to see what's happening in novice teachers classrooms and provide support? The simple answer is we don't know what the impact of that will be, but that's a problem that needs to be solved somehow. So why don't we test the technology and evaluate what the impact is? Because there could be great gains, but there could be unintended consequences which cancel out those great gains.
10:13
Manuela Perteghella
-
-
Manuela Perteghella
And that brings me to my next question, actually, what safeguards would you like to see? Would the panel would like to see to ensure that AI and tech support, as their supportive role in the classroom, rather than undermine teachers professional judgement expertise and also reduce the role of of the teacher in the classroom to inspire the critical thinking skills to students? So yes. I
10:14
Rebecca Paul
-
-
Rebecca Paul
think, I mean, there's a number of safeguards and pieces across that piece, but the one that I think is one that's a quick win. If we look to China, they made a very bold legislation on the role of Agentic AI inside all edtech products for minors. And that means that they have banned the use of human like personas and avatars. And that's a really subtle but important distinction. That means that young people aren't going to start forming relationships with agents that aren't humans in ways that we know are problematic in later years.
So I think there's some interesting ways of thinking about different parts of the piece. But then across all of it, we've got data governance, data sources, data security, interoperability, lots of things that impact the way data flows in and out of these different AI tools.
So I think there's some interesting ways of thinking about different parts of the piece. But then across all of it, we've got data governance, data sources, data security, interoperability, lots of things that impact the way data flows in and out of these different AI tools.
10:15
Speaker 1
-
Speaker 1
So I think also we need to encourage a lot of precision about precisely what we're talking about when we're talking about AI, when we're talking about technology, what we're rejecting, what we're accepting. So that idea of really thinking about specific affordances that we can either find acceptable or not within education is really important. So, you know, it might well be that certain kinds of AI that use very data extractive practices that cause a lot of privacy concerns or technologies that it's really almost impossible.
It is impossible to understand the level of bias or inaccuracies. You know, we want to be very precise about the kinds of things we are unhappy about with technology and the kinds of things we want to protect and design for, and alongside other things that we've talked about already today. I do think expanding the kinds of evidence that we see is important to not just think about the kinds of outcomes of particular educational interventions using technology, but also these processes, these unintended consequences, these but these range of other things.
That means when we use technology, what does that mean to the experience of young people engaging in education? And I think when we have a stronger way of accounting for teacher knowledge and a way of thinking about evidence, then we can start to sort of shift the procurement requirements of schools and so on. That starts to sort of change the demand structure around technologies, and starts to sort of shift the market in ways that we think are important. Therefore, safeguarding the things that we think are educationally important.
It is impossible to understand the level of bias or inaccuracies. You know, we want to be very precise about the kinds of things we are unhappy about with technology and the kinds of things we want to protect and design for, and alongside other things that we've talked about already today. I do think expanding the kinds of evidence that we see is important to not just think about the kinds of outcomes of particular educational interventions using technology, but also these processes, these unintended consequences, these but these range of other things.
That means when we use technology, what does that mean to the experience of young people engaging in education? And I think when we have a stronger way of accounting for teacher knowledge and a way of thinking about evidence, then we can start to sort of shift the procurement requirements of schools and so on. That starts to sort of change the demand structure around technologies, and starts to sort of shift the market in ways that we think are important. Therefore, safeguarding the things that we think are educationally important.
10:16
Speaker 2
-
Speaker 2
And I think that that point is really important about schools being, if you like, informed consumers of the technology that they buy, and there is a real risk that packages are being pushed on schools, very hard, expensive packages. We've got finite resources in schools, and that those investments will not be as effective or as safe as we would want them to be. I'd agree with what my colleagues have said, just maybe to add two more things. One is something that concerns us a lot is around the interaction between AI data protection and privacy.
And again, we're not confident that trusts and schools across the system really understand those data protection and privacy risks that can be attached to the use of AI, the uploading of information and data that may contravene GDPR. So I think part of the training and support that we're thinking about here has to involve how you use AI in ways that respect, if you're going to use it, respect the data protection and privacy rights of staff and of learners as well.
I think the fundamental principle in terms of safety that we would want to see applied is that pedagogical primacy, the idea that technology, when it is used, supports teaching and learning and supports the professional decisions that teacher makes and doesn't supplant them. That seems to me to be a really important safeguard in terms of, among other things, the quality of children's educational experiences.
And again, we're not confident that trusts and schools across the system really understand those data protection and privacy risks that can be attached to the use of AI, the uploading of information and data that may contravene GDPR. So I think part of the training and support that we're thinking about here has to involve how you use AI in ways that respect, if you're going to use it, respect the data protection and privacy rights of staff and of learners as well.
I think the fundamental principle in terms of safety that we would want to see applied is that pedagogical primacy, the idea that technology, when it is used, supports teaching and learning and supports the professional decisions that teacher makes and doesn't supplant them. That seems to me to be a really important safeguard in terms of, among other things, the quality of children's educational experiences.
10:18
10:18
Speaker 3
-
Speaker 3
Absolutely. The the prominent cognitive scientist Dan Willingham spoke wrote recently around this topic. And he had a very simple principle, which I think is useful just to say out loud. And he said, we need to ensure that AI tools aren't used for tasks wherein the user would benefit from doing the mental work themselves. And if we have a guiding principle as simple as that, it might help shape the work we do. And picking up on Darren's point, the simple answer here in terms of safeguarding is ensuring the quality of training.
We have put a massive. We have a massive burden is being put onto schools to try and reconcile and understand what on earth to do with this technology. They have found money from unbelievably finite small pots to create. I champion roles. They've got colleagues who are working unpaid, above and beyond to try and upskill and understand this technology and think about what that means for their pupils. We need to support the sector and ensure every teacher in this country has access to high quality, evidence informed training, not just in, and definitely not just in how to use the tools, but in how the tools work, the ethics of use, and the considerations around cognitive offloading.
Because we want to upskill our profession to make these choices themselves informed, intentional evidence led decisions. And there is an opportunity to do that. We're lucky in this country. We have national statutory frameworks, but we have nothing in those frameworks that talks about AI at all. And urgent update to those frameworks would ensure that there are many thousands of teachers already accessing professional development in this country, would immediately start to get high quality AI training, rather than leaving it largely to tech companies, which obviously providing that training from a particular angle.
We have put a massive. We have a massive burden is being put onto schools to try and reconcile and understand what on earth to do with this technology. They have found money from unbelievably finite small pots to create. I champion roles. They've got colleagues who are working unpaid, above and beyond to try and upskill and understand this technology and think about what that means for their pupils. We need to support the sector and ensure every teacher in this country has access to high quality, evidence informed training, not just in, and definitely not just in how to use the tools, but in how the tools work, the ethics of use, and the considerations around cognitive offloading.
Because we want to upskill our profession to make these choices themselves informed, intentional evidence led decisions. And there is an opportunity to do that. We're lucky in this country. We have national statutory frameworks, but we have nothing in those frameworks that talks about AI at all. And urgent update to those frameworks would ensure that there are many thousands of teachers already accessing professional development in this country, would immediately start to get high quality AI training, rather than leaving it largely to tech companies, which obviously providing that training from a particular angle.
10:19
Manuela Perteghella
-
-
Manuela Perteghella
And my last question, so it was fundamental for me as a lecturer to plan my lectures and in particular my seminars, because I knew my students, I knew the class dynamics and I knew their needs. So it wasn't just an admin task. It was the really the, the, the foundations to then foster that critical thinking and that share knowledge in, in the classroom. Are there particular areas of teaching where AI should play only a limited role, regardless of how capable the technology becomes?
10:20
Speaker 3
-
Speaker 3
I think so, absolutely. And I think the particular challenge this makes me think about is the novice new teacher versus the experienced teacher, and the absolute challenge head teachers have at the moment to try and think of a policy that works for both colleagues. So we know that helping an expert to work more effectively is not the same thing as helping a novice become an expert. And how on earth we're meant to have an all encompassing policy in a school for somebody that needs to do the hard yards, cognitive grapple of learning how to plan, because we know that is the foundational to being a great teacher.
But yet we want our 20 year, 15 year experience colleague to experiment with how it might enhance the quality of their lesson planning, how on earth we wrestle with that. That's a real tricky situation for the sector to try and pick at the moment.
But yet we want our 20 year, 15 year experience colleague to experiment with how it might enhance the quality of their lesson planning, how on earth we wrestle with that. That's a real tricky situation for the sector to try and pick at the moment.
10:21
Speaker 2
-
Speaker 2
I think. I think that's a that's right. You're absolutely right about planning as well. A lesson plan is a is a product. And planning a lesson is a kind of professional process that a teacher goes through. And it's very important. That point, I think about maybe teachers coming through initial teacher training now, maybe not having those experiences, particularly when they're in schools of thinking about planning and doing that planning as a kind of cognitive professional exercise.
That is a concern. That's not just AI. However, we certainly hear from student teachers who go on school experience what used to be called teaching practice when I was learning to training to teach. When they're in schools, they're deployed in schools where the plans are simply given to them to follow. And we know that some initial teacher training providers are very concerned that, among other things, to meet the standards to be awarded Qprts, you've got to be able to demonstrate that you can plan lessons.
And if the school you're in is handing you lessons and basically telling you to read that off a script. That's concerning. AI, I think, just adds to that concern , because clearly, if you're just presented with something that AI has churned out, you're not going through the process that you described that is so important to high quality provision.
That is a concern. That's not just AI. However, we certainly hear from student teachers who go on school experience what used to be called teaching practice when I was learning to training to teach. When they're in schools, they're deployed in schools where the plans are simply given to them to follow. And we know that some initial teacher training providers are very concerned that, among other things, to meet the standards to be awarded Qprts, you've got to be able to demonstrate that you can plan lessons.
And if the school you're in is handing you lessons and basically telling you to read that off a script. That's concerning. AI, I think, just adds to that concern , because clearly, if you're just presented with something that AI has churned out, you're not going through the process that you described that is so important to high quality provision.
10:22
Speaker 1
-
Speaker 1
Yes, I would echo all of those comments, and it's definitely something that we've seen in our research. People are very concerned about AI for for planning lessons, particularly for less experienced teachers and what that might mean. And so I think really thinking about this intersection between expertise and AI is going to be crucial. But also, as you say, I think when we're thinking about kind of encouraging people into the profession and encouraging people to stay there, I think we really have to think about what's worthwhile to people, what's meaningful to people in doing their job well, and therefore thinking more carefully about how we use AI as part of that process.
10:22
10:22
Rebecca Paul
-
-
Rebecca Paul
involved in two studies that are. The EEF is currently conducting teacher choice studies, ones looking at the use of copilot to support lesson planning. And the other one is looking at a particular EdTech product that is there to support assessing in specific marking student work. And I think one of the red lines that in providing guidance for teachers is around grading. So currently sitting in different institution policies, some are very hardline in that you cannot use an AI to actually grade students work, particularly in anything that has a summative piece of coursework, for example.
But thinking about where it sits in the process of assessing, is it a red line that you should always have reviewed the work yourself first before putting it into an AI engine to take that task on? I think these are where we've got to really understand where the benefits are. The allure is there, and I can see many teachers looking to cut time by fully automating those processes, but all of the same points around lesson planning have to go alongside the process of assessing.
It's critical to a teacher or lecturer deciding next steps for that group of students. So I think these we've got to look really judiciously at how we embed some of these tools. If we see there's opportunities.
But thinking about where it sits in the process of assessing, is it a red line that you should always have reviewed the work yourself first before putting it into an AI engine to take that task on? I think these are where we've got to really understand where the benefits are. The allure is there, and I can see many teachers looking to cut time by fully automating those processes, but all of the same points around lesson planning have to go alongside the process of assessing.
It's critical to a teacher or lecturer deciding next steps for that group of students. So I think these we've got to look really judiciously at how we embed some of these tools. If we see there's opportunities.
10:24
10:24
Sureena Brackenridge
-
-
Sureena Brackenridge
Thank you. So following on from those points, the government has said that. AI can reduce teacher workload and automating routine tasks. But will it impact on wellbeing and is it a sustained reduction of workload? Because we've also heard evidence that it may just be a reconfiguration of workload, with additional responsibilities put onto teachers? So how strong is the evidence that AI is reducing teacher workload in practice? And I'll start with Darren.
10:24
Speaker 2
-
Speaker 2
Thank you. I don't think the evidence is very strong at all that it's reducing teacher workload. There's a kind of concept, if you like, of task efficiency. So can a task be done more quickly. But we say that our members earlier this year and over around three quarters had used AI in one form or another in their work, but only 7% had said it had reduced workload. So we need to explore why so few teachers are actually reporting that overall, the use of AI isn't reducing their workload.
And I think there could be a number of reasons for that. One of them is, is that just using AI in and of itself creates additional workload burdens. So for example, if you're using AI to support planning, that can be very helpful. But if you're doing that appropriately, you're then going to check all the citations, all the evidence, perhaps that the AI platform has produced in developing that lesson plan. So you're kind of, as you said, transforming workloads.
So it takes another form, but the overall amount of workload isn't reduced. I think another issue, and maybe this goes to the point we were just discussing, is I think many teachers are very wary if they have the choice. And we know that some don't. But most teachers who have the choice are very wary of just using an AI platform to generate a lesson plan, which they then deliver, that seems to be sold as the biggest workload reduction benefit.
But for very good professional reasons. Teachers are nervous about doing that for all the reasons that we've discussed. So I think the jury is very much still out on the extent to which AI currently reduces workload, and also the extent to which it will, in the long term, lead to sustainable workload reductions for teachers.
And I think there could be a number of reasons for that. One of them is, is that just using AI in and of itself creates additional workload burdens. So for example, if you're using AI to support planning, that can be very helpful. But if you're doing that appropriately, you're then going to check all the citations, all the evidence, perhaps that the AI platform has produced in developing that lesson plan. So you're kind of, as you said, transforming workloads.
So it takes another form, but the overall amount of workload isn't reduced. I think another issue, and maybe this goes to the point we were just discussing, is I think many teachers are very wary if they have the choice. And we know that some don't. But most teachers who have the choice are very wary of just using an AI platform to generate a lesson plan, which they then deliver, that seems to be sold as the biggest workload reduction benefit.
But for very good professional reasons. Teachers are nervous about doing that for all the reasons that we've discussed. So I think the jury is very much still out on the extent to which AI currently reduces workload, and also the extent to which it will, in the long term, lead to sustainable workload reductions for teachers.
10:26
10:26
Speaker 3
-
Speaker 3
Yeah, I would very much agree . There is limited evidence in this space and an urgency to do more evaluation. There was one research trial done by the EEF, which looked at about 259 secondary school teachers on their year seven and eight marking, and they found that those colleagues using AI to help plan their lessons saved 25 minutes a week. And what they did with that 25 minutes wasn't part of the trial. So there really is a much, much greater need to delve deeper into this area to understand.
I think one of the reasons that's worth exploring is it's the lack of training that is meaning when people are using AI, they are maybe not using it in a judicious way. So if you don't understand how it works, then you're going to maybe ask it to do the wrong kind of tasks that don't actually help you. But to your earlier point, you made in your the start of your question in terms of back office function, we have spoken to many schools who are doing brilliant, innovative things with back office, non pupil facing functions that are speeding up administration tasks and efficiencies.
There again, we don't know what the impact is in terms of what is that time saved turn into in terms of pupil impact. And our North star has to always be what's the impact on pupils. So I would just very much agree with what Darren said. And this urgency again, which you'll get bored of me saying, is we need to do desperately more evaluation in this space urgently.
I think one of the reasons that's worth exploring is it's the lack of training that is meaning when people are using AI, they are maybe not using it in a judicious way. So if you don't understand how it works, then you're going to maybe ask it to do the wrong kind of tasks that don't actually help you. But to your earlier point, you made in your the start of your question in terms of back office function, we have spoken to many schools who are doing brilliant, innovative things with back office, non pupil facing functions that are speeding up administration tasks and efficiencies.
There again, we don't know what the impact is in terms of what is that time saved turn into in terms of pupil impact. And our North star has to always be what's the impact on pupils. So I would just very much agree with what Darren said. And this urgency again, which you'll get bored of me saying, is we need to do desperately more evaluation in this space urgently.
10:27
10:27
Speaker 1
-
Speaker 1
Yes. So I would also very much support a reconfiguring of teachers work. I'm not sure it's ever going to save anybody any time. I mean, does email save us time? I don't feel like it does. So technology's often promised this tech saving, but in our research we very much saw teachers saving time to some degree using it for very repetitive or non-teaching tasks. Maybe a bit of lesson planning, particularly for lessons that aren't their core responsibility.
So when they're covering lessons, for example, but it's also taking time, you know, it's taking time to think about, well, how do we teach to ensure we don't have cognitive shortcuts? How do we manage all of the tech and the distraction in the classroom? How do we manage all of the laptops and so on. So it's, it's kind of giving and taking time at the same in the same moment. So I think reconfiguring sort of allows us to see the complexity of when we put technology in work together.
I think one of the challenges with the research in this space is not just that we haven't got enough of it. I also think it's because it's very hard for people to measure time. And it's also conceptually, for example, in that study, which is a good quality study, we're looking at the quality of the lesson plan, but we're not looking at the implications on the actual lesson practice. So even if it saves time, we don't know the kind of downstream effects of that.
So yeah, it's a complicated space.
So when they're covering lessons, for example, but it's also taking time, you know, it's taking time to think about, well, how do we teach to ensure we don't have cognitive shortcuts? How do we manage all of the tech and the distraction in the classroom? How do we manage all of the laptops and so on. So it's, it's kind of giving and taking time at the same in the same moment. So I think reconfiguring sort of allows us to see the complexity of when we put technology in work together.
I think one of the challenges with the research in this space is not just that we haven't got enough of it. I also think it's because it's very hard for people to measure time. And it's also conceptually, for example, in that study, which is a good quality study, we're looking at the quality of the lesson plan, but we're not looking at the implications on the actual lesson practice. So even if it saves time, we don't know the kind of downstream effects of that.
So yeah, it's a complicated space.
10:29
10:29
10:29
10:29
10:29
Speaker 2
-
Speaker 2
I think I'd pick up pick up on what Pip said around the potential, around what you might call back office functions and around some of the administrative and clerical burdens that teachers have. That might be an area to explore, because I think in terms of some of the more high stakes things we've been discussing, the stakes there are lower. So those might be things to to consider. It still has to be done extremely carefully. I mean, I go back to this point around schools, trusts, employers, not really understanding things around data protection and privacy.
So there has to be a lot of work done to upskill and enhance the knowledge of schools and employers on that. But I think that might be an area where we can focus our attention. I think some of the higher stakes stuff, particularly that stuff around pupil teacher interactions, I think we need a lot more. As colleagues have said, we need a lot more evidence about how that can be done safely and effectively. So there are some opportunities, but I think we still need to be quite cautious.
So there has to be a lot of work done to upskill and enhance the knowledge of schools and employers on that. But I think that might be an area where we can focus our attention. I think some of the higher stakes stuff, particularly that stuff around pupil teacher interactions, I think we need a lot more. As colleagues have said, we need a lot more evidence about how that can be done safely and effectively. So there are some opportunities, but I think we still need to be quite cautious.
10:30
Speaker 3
-
Speaker 3
I think marking actually is one opportunity that is worth examining in terms of if it's done well and judiciously, and it's a complement to the human LED marking the potentially as a role there, certain subjects. The marking burden is huge. As an English teacher, I can speak to how long it takes to mark a class of 30 mock exams and if there's an opportunity to use some of the really good technology out there that's using comparative judgements along with AI, then actually what some of the early studies are finding is there is a reduction in in some biases, and there's actually more consistency in the distribution of marks.
But again, we just need to test that and to see what are the consequences of introducing judicious deployment of AI, marking where it can be an aid to teacher and it frees them up. And it comes back to the point, though, what does the teacher do with that free time? What do we imagine is going to suddenly happen with this free time? If there's nothing put in place in terms of what we want, is that a chance for increased professional development? Is that an opportunity to rehearse and practice lessons more? We're not going to see the gains land on our pupils.
But again, we just need to test that and to see what are the consequences of introducing judicious deployment of AI, marking where it can be an aid to teacher and it frees them up. And it comes back to the point, though, what does the teacher do with that free time? What do we imagine is going to suddenly happen with this free time? If there's nothing put in place in terms of what we want, is that a chance for increased professional development? Is that an opportunity to rehearse and practice lessons more? We're not going to see the gains land on our pupils.
10:31
10:31
10:31
Rebecca Paul
-
-
Rebecca Paul
would bring in the my experience of working with the sector at this point, actually, because one of the things that in working alongside edtech companies, some of them at the early stages of developing their different tools and services, and others really quite the big guys. One of the things they're all unclear of is what the impact goals for the technologies that they're building actually are. They're there to sell, you know, sell a hype really, and people are buying into it.
I think there's the opportunities that the new DfE EdTech impact testbed project is giving us is worth recognising here, because what this is going to enable us to do is work more closely with the tech partners and the schools and practitioners, in part to generate the much needed evidence. But also it is about understanding those perspectives which are different. There's different goals involved. We have to be the guardians of the for the pupils and the system.
And that's where guardrails and legislation comes in, because we're much closer to seeing what the problems are when we're working alongside the developers, rather than always seeing their tools as something to be externally evaluated. So that test and learn model, that's going to allow us to take promising practices from schools and colleges, work with researchers and the companies to really hone in on what is the impact we're looking for here. If it's workload, what aspect of workload are we talking about? And we'll have a much finer body of evidence, hopefully at more scale to be able to draw on, and I'd love to see that project be sustained long after the funding is going.
Can we change the culture in our schools and colleges to enable them to have this sort of evaluation mindset on any innovation? It's AI now, you know, it was the whiteboards 20 years ago, but we we've we've seen these cycles regularly and we've never quite found the right school level and college level processes that are manageable to help us do that evidence generation at home, at home base.
I think there's the opportunities that the new DfE EdTech impact testbed project is giving us is worth recognising here, because what this is going to enable us to do is work more closely with the tech partners and the schools and practitioners, in part to generate the much needed evidence. But also it is about understanding those perspectives which are different. There's different goals involved. We have to be the guardians of the for the pupils and the system.
And that's where guardrails and legislation comes in, because we're much closer to seeing what the problems are when we're working alongside the developers, rather than always seeing their tools as something to be externally evaluated. So that test and learn model, that's going to allow us to take promising practices from schools and colleges, work with researchers and the companies to really hone in on what is the impact we're looking for here. If it's workload, what aspect of workload are we talking about? And we'll have a much finer body of evidence, hopefully at more scale to be able to draw on, and I'd love to see that project be sustained long after the funding is going.
Can we change the culture in our schools and colleges to enable them to have this sort of evaluation mindset on any innovation? It's AI now, you know, it was the whiteboards 20 years ago, but we we've we've seen these cycles regularly and we've never quite found the right school level and college level processes that are manageable to help us do that evidence generation at home, at home base.
10:33
Speaker 1
-
Speaker 1
And just to add briefly, though, I would still campaign for a way of thinking about this as reconfiguring work and not saving time. I think that then encourages us to think about the meaningful work that teachers do, and trying to increase that through technology, rather than this kind of efficiency saving, time saving, which I don't think. I don't think work is like a cake, that we can kind of cut up into bits and save a quarter.
I think it's something that happens in a much more complicated way. And so thinking more about meaningful work, I think would be much more useful for the sector.
I think it's something that happens in a much more complicated way. And so thinking more about meaningful work, I think would be much more useful for the sector.
10:34
10:34
Speaker 1
-
Speaker 1
Yeah. So so most of my work is primarily in secondary schools. So it's a little bit difficult for me to to answer that question. But I do think it's. You know, a, the, the nature of teachers work in secondary is obviously quite different. And therefore, I think this possibility for reconfiguring is maybe greater.
10:34
Speaker 3
-
Speaker 3
I would just say that I think if we're thinking about the teacher use of AI, it's really irrelevant whether it's primary or secondary. It's still. If we keep focus on the principle of are they using it to support their development of expertise, or are we using it as a shortcut to replace it? And that was our guiding principle. I think we could do some some quite good work if we kept that front of mind.
10:35
10:35
Rebecca Paul
-
-
Rebecca Paul
Yeah, I think it's where we place the educating about AI. You know, I mean, even if we decide to to steer clear for it, we still need to educate our young people about it. And how we do that in the earlier stages looks very different to the later stages. And ultimately, we want young people to come out able to program it. So we've got a huge need there in terms of the upskilling, thinking about how that education reform is going to allow for us to upskill our population to live in the world that lies ahead of us.
10:36
10:36
Speaker 3
-
Speaker 3
Can I just add sorry on that point as well. That's okay. I think given thinking about the age that we're living in now, and I think it's just worth thinking about AI in that context, we're living in this time of information system change. And when our information systems change, we change as a society. You know, it happened before the invention of writing and the invention of the printing press, and it's happening again now. We're living through this moment right now, and that teacher pupil relationship is becoming ever more important in this particular age where we see trust is decreasing.
We have an abundance of information, but we are getting this increasing scarcity, scarcity of trust. So I think whether you are primary or secondary, a core focus for our teachers is giving them the support to focus on that interpersonal relationship development and how we can build those relationships with our pupils in this particular information climate we're living in. And the good news here is that pupils generally trust their teachers, particularly at primary. So there's something interesting to think about at primary in terms of being really careful about how AI possibly could be an interference in that really beautiful teacher pupil relationship that sets young people up to go on to do great things.
We have an abundance of information, but we are getting this increasing scarcity, scarcity of trust. So I think whether you are primary or secondary, a core focus for our teachers is giving them the support to focus on that interpersonal relationship development and how we can build those relationships with our pupils in this particular information climate we're living in. And the good news here is that pupils generally trust their teachers, particularly at primary. So there's something interesting to think about at primary in terms of being really careful about how AI possibly could be an interference in that really beautiful teacher pupil relationship that sets young people up to go on to do great things.
10:37
10:37
10:37
10:37
Speaker 2
-
Speaker 2
I think it's about exploring on what basis the government holds that belief. I think there are two kind of strands to that. One I'm much more skeptical about than the other. The other is, is the first one, I guess is workload. So as we've just been discussing, the government's narrative seems to be that AI can exert downward pressure on workload. And we know that workload is the biggest driver of exit from the profession.
And it also does, I think, deter lots of graduates, graduates from thinking about teaching as a career. But I don't think the case is proven that the increasing use of AI is going to reduce workload in a way that supports addressing some of those retention challenges that we face. I think the other question, maybe a more interesting area to explore is perhaps what the government is thinking is that perhaps people coming into the workplace now, graduates coming into the labour market now would expect to have AI tools available and would want those to be used as part of their work.
So that becomes part of the expectation that that's reasonable. But I don't think any of that addresses those fundamental drivers of the recruitment and retention crisis that are really well established. We know that workload, teacher wellbeing, pupil behaviour, we know that those are cited, frankly, year after year by teachers and leaders as why there is such a challenge in respect of recruitment and retention of teachers in the profession. So AI may have a role to play, but I think at the moment, the safest thing we can say is that that is going to be at best at the margins, and there are root causes of that crisis that we need to and the government needs to address.
And it also does, I think, deter lots of graduates, graduates from thinking about teaching as a career. But I don't think the case is proven that the increasing use of AI is going to reduce workload in a way that supports addressing some of those retention challenges that we face. I think the other question, maybe a more interesting area to explore is perhaps what the government is thinking is that perhaps people coming into the workplace now, graduates coming into the labour market now would expect to have AI tools available and would want those to be used as part of their work.
So that becomes part of the expectation that that's reasonable. But I don't think any of that addresses those fundamental drivers of the recruitment and retention crisis that are really well established. We know that workload, teacher wellbeing, pupil behaviour, we know that those are cited, frankly, year after year by teachers and leaders as why there is such a challenge in respect of recruitment and retention of teachers in the profession. So AI may have a role to play, but I think at the moment, the safest thing we can say is that that is going to be at best at the margins, and there are root causes of that crisis that we need to and the government needs to address.
10:39
Rebecca Paul
-
-
Rebecca Paul
Yeah, I think I mean, I tend to think more about the incentives than the challenges and the issues and look back to some of the incentives that were around that have successfully helped teachers to stay in the profession. So the idea, I mean, we heard mention about AI champions and digital champions recognising in the system the value of these new people in the education workforce that we need in our schools now. So whether it's education data analysts, whether it's education, AI specialists, but there's this new leadership roles that that don't necessarily fit in the existing structures that are going to help us get on top and stay ahead thinking about what that looks like in the professional routes for teachers.
So in some ways, opening up opportunities to stay because often the workload, it's the same old, same old, you don't necessarily have interesting progression routes within the system as it is now. There's an opportunity there, most certainly, and to, to see how that aligns with teachers being able to gain masters credits or do something professionally rewarding that, you know, often inevitably take them out of the classroom, but they stay in the system.
So in some ways, opening up opportunities to stay because often the workload, it's the same old, same old, you don't necessarily have interesting progression routes within the system as it is now. There's an opportunity there, most certainly, and to, to see how that aligns with teachers being able to gain masters credits or do something professionally rewarding that, you know, often inevitably take them out of the classroom, but they stay in the system.
10:40
10:40
10:40
Speaker 3
-
Speaker 3
I think, in a sort of odd way, it might be that we see an increase in the number of people wanting to become teachers, because it becomes one of the professions where there is still a lot of human interaction. And for people where that work appeals and it is AI largely free may become a massive plus point for teaching. We may actually see a sort of a side benefit from AI in terms of teacher recruitment in that way and retention.
And one thing that I think is important to say from talking to teachers is that we often, and rightly so, are focused on evaluating efficacy and and impact on young people. But there is also something important in terms of how does AI make teachers feel about being a teacher? And that sounds a bit soft and woolly, but it's really important in terms of how do I feel as a teacher? If you take away my lesson planning, which is my chance for creativity, and it's my ability to think hard about my pupils, do I still want to be a teacher , just delivering something somebody else has created that might appeal to some people that wouldn't have thought about teaching, but I think we would see a lot of people who currently teach well suddenly find that career path much less desirable.
And one thing that I think is important to say from talking to teachers is that we often, and rightly so, are focused on evaluating efficacy and and impact on young people. But there is also something important in terms of how does AI make teachers feel about being a teacher? And that sounds a bit soft and woolly, but it's really important in terms of how do I feel as a teacher? If you take away my lesson planning, which is my chance for creativity, and it's my ability to think hard about my pupils, do I still want to be a teacher , just delivering something somebody else has created that might appeal to some people that wouldn't have thought about teaching, but I think we would see a lot of people who currently teach well suddenly find that career path much less desirable.
10:41
10:41
Speaker 1
-
Speaker 1
And I agree with with all of the things that have been said. And I think that's why I'm sort of arguing for this idea of meaningful work and a reconfiguring of work rather than efficiency gains, because it's it's creating meaningful work that will attract and retain people to the profession. I think another aspect of this is thinking about public perceptions of teaching, which is really important. And I think there's some really interesting tensions there that are. Research has shown.
So teachers have talked to us about how they're concerned that the public might have, you know, conflicting ideas of when they use AI. So on the one hand, they're kind of forward thinking and modern and so on. But on the other hand, are they being lazy if they use AI to create reports or do they not have expertise if they're using AI to generate their lesson planning? So this is kind of tension in terms of how the public are also thinking about AI, and then how they're thinking about the profession, which may well have implications for people being attracted to it.
So teachers have talked to us about how they're concerned that the public might have, you know, conflicting ideas of when they use AI. So on the one hand, they're kind of forward thinking and modern and so on. But on the other hand, are they being lazy if they use AI to create reports or do they not have expertise if they're using AI to generate their lesson planning? So this is kind of tension in terms of how the public are also thinking about AI, and then how they're thinking about the profession, which may well have implications for people being attracted to it.
10:42
10:42
Speaker 3
-
Speaker 3
Very, very little, very little. It's theoretical at the moment. In terms of the evidence, there is some studies that have been done recently, 2024, but this was looking at students on degree courses and how AI was used to help identify those who were underperforming and likely to drop out, and then targeted support could be put in place. As far as we understand, there's been no testing done with teachers specifically to see how that could be done, but theoretically it could be.
However, it opens up a whole host of issues in terms of that level of monitoring of of a student or a teacher's performance.
However, it opens up a whole host of issues in terms of that level of monitoring of of a student or a teacher's performance.
10:43
10:43
10:43
Rebecca Paul
-
-
Rebecca Paul
10:44
Rebecca Paul
-
-
Rebecca Paul
It's always going. I mean, it's always there's a set of competencies around AI that exist that are a good starting point. These have been developed by Unesco as an international set of competencies. So they outline, you know, quite generally. But making sense of them is in the in the practice. So it's mentorship support and in the school setting at all stages of a teacher's professional career to continue to grow and learn.
10:44
Speaker 3
-
Speaker 3
We have a real problem that if we if we don't train the mentors and the experienced colleagues, we only train the trainees, we're going to get a divergence and not coming to cumulative coming together in terms of what AI training should should contain. The one thing that we see missing from most provision that's out there, provided by my tech companies, is there's nothing on the social and psychological effects of AI use. It's not talked about in the broader landscape of the changing digital information age that we live in.
There's very little attendance to the cognitive science of how learning happens in relation to what we know about how LLMs work, and there's probabilistic inference. And those two things together we found from our courses help the light bulb sort of moment come for trainees at thinking. This is why AI might limit my own learning. And actually, maybe it could be useful in this little bit of use over here. But unless we give people that true understanding of how they work, how the science of learning works, the social and psychological impacts of of AI, we're never going to get that well informed generation of teachers who can make intentional choices.
You can't do that in a 30 minute session. All the HCI providers and providers across the country are working their socks off to upskill themselves and find space in the margins to deliver 30 minutes here, maybe an hour, maybe two hours. We need to just change the frameworks to make an absolute statutory entitlement that every trainee teacher, every every MP participant in this country has access to high quality, evidence informed AI training. And that will be a really good step to safeguarding our pupils and our teachers.
There's very little attendance to the cognitive science of how learning happens in relation to what we know about how LLMs work, and there's probabilistic inference. And those two things together we found from our courses help the light bulb sort of moment come for trainees at thinking. This is why AI might limit my own learning. And actually, maybe it could be useful in this little bit of use over here. But unless we give people that true understanding of how they work, how the science of learning works, the social and psychological impacts of of AI, we're never going to get that well informed generation of teachers who can make intentional choices.
You can't do that in a 30 minute session. All the HCI providers and providers across the country are working their socks off to upskill themselves and find space in the margins to deliver 30 minutes here, maybe an hour, maybe two hours. We need to just change the frameworks to make an absolute statutory entitlement that every trainee teacher, every every MP participant in this country has access to high quality, evidence informed AI training. And that will be a really good step to safeguarding our pupils and our teachers.
10:46
10:46
Speaker 2
-
Speaker 2
I don't think they are. And I think this is where the dfe's role is really important. We're asking schools, sometimes schools on their own, not part of trusts or not with local authority support, to make really complex decisions around how AI is deployed in their teaching and learning by their staff and, and in respect of their pupils. So I think there is a massive issue there around confidence. And I think the DfE has a role to play in that much of what the DfE has produced, some of it I think is quite interesting and quite helpful, but it seems quite laissez faire.
So what it's saying to schools is, well, you should do this or you could do this. I think what schools are looking for are much clearer rules of the road about AI. AI, if it is going to be integrated into teaching and learning and in other functions within schools, how that is best done and what the dfe's expectations are around that. Just just briefly picking up your earlier point about about training, I think across the piece, our members are profoundly dissatisfied with the CPD.
They experience in all areas and in all respects. And I think part of that is because of excessive workload. I think part of that is also because CPD is marginalised into small spaces. That's what Pip's organisation exists to, to try and and address. The other point I would make about the training that teachers actually experience, and this goes back to a point we were discussing earlier, is a lot of that comes from the suppliers of the AI platforms themselves, either the ones that are being used, and maybe there's a case for that, but the ones that are being pitched at schools.
So there's a kind of issue there around the objectivity, perhaps, of some of the advice that schools are receiving, because if it's coming from an organisation that's attempting to sell products into schools, I can't speak for all teachers. But as a teacher, if I were sitting in a staff room listening to that, I'd have a little bit of skepticism about what I was being told and why. Does
So what it's saying to schools is, well, you should do this or you could do this. I think what schools are looking for are much clearer rules of the road about AI. AI, if it is going to be integrated into teaching and learning and in other functions within schools, how that is best done and what the dfe's expectations are around that. Just just briefly picking up your earlier point about about training, I think across the piece, our members are profoundly dissatisfied with the CPD.
They experience in all areas and in all respects. And I think part of that is because of excessive workload. I think part of that is also because CPD is marginalised into small spaces. That's what Pip's organisation exists to, to try and and address. The other point I would make about the training that teachers actually experience, and this goes back to a point we were discussing earlier, is a lot of that comes from the suppliers of the AI platforms themselves, either the ones that are being used, and maybe there's a case for that, but the ones that are being pitched at schools.
So there's a kind of issue there around the objectivity, perhaps, of some of the advice that schools are receiving, because if it's coming from an organisation that's attempting to sell products into schools, I can't speak for all teachers. But as a teacher, if I were sitting in a staff room listening to that, I'd have a little bit of skepticism about what I was being told and why. Does
10:48
10:48
10:48
10:48
10:48
10:48
Rebecca Paul
-
-
Rebecca Paul
Leave her on that last point there around working with the industry because, you know, we we can set in frameworks for how we want edtech tools to look in this country for our learners. We can also set in processes. So even if you are selling, do you sell with a pilot evaluation toolkit? That means you provide some guidance on the specifics of the impacts you allow the schools and colleges to refine it, but you give them a little instrument or tool to help them, then decide if they go and procure.
So these sorts of approaches that mean you don't have to make such big decisions with no evidence, you can essentially, even if it's been evidenced elsewhere, re pilot it in your setting and then make a decision.
So these sorts of approaches that mean you don't have to make such big decisions with no evidence, you can essentially, even if it's been evidenced elsewhere, re pilot it in your setting and then make a decision.
10:48
Speaker 3
-
Speaker 3
It's very quickly build on that point to say as well and advocate for when we say working with the tech companies, we need to make sure teachers are compensated to do that. Currently, it's often, oh, here's an opportunity to come and work the tech company and have some input on the tool. And they're doing above and beyond in their own free time, or their school is covering them by paying a supply teacher to take that lesson. And ultimately, who's going to benefit from the product, hopefully the sector.
But we also know other people are going to benefit. So unless we start compensating our teachers properly to be true partners, it's always going to be a very one sided relationship.
But we also know other people are going to benefit. So unless we start compensating our teachers properly to be true partners, it's always going to be a very one sided relationship.
10:49
10:49
Speaker 1
-
Speaker 1
And just very briefly, I think it's really important for us to start to capture that expertise from teachers and to really sort of build it and grow it, and to encourage kind of participatory forms of design that aren't quite kind of tokenistic or too quick. So really thinking while we're thinking about training, thinking about culture change and thinking about how do we learn from each other and share that knowledge and build that knowledge to shape the industry?
10:49
10:49
Peter Swallow
-
-
Peter Swallow
Darren Northcott, I think you you've been clear without wanting to put words into your mouth, that that government needs to step in and support schools to establish clear policies governing the use of AI by staff and pupils. You said previously as well that schools don't understand data privacy concerns. I wondered if anyone wanted to sort of take take the counterfactual on that point. Is the risk of too much decision making being taken centrally by government on what the the policy should be governing this, that schools aren't going to be in a position to schools and colleges aren't going to be in a position to innovate, to adapt, to apply innovative uses of AI to their specific settings.
10:50
10:50
Speaker 3
-
Speaker 3
Agree with Darren. I think, I think it's difficult, I think I mean, it's interesting, the Council of Europe just published a set of AI recommendations. 28 of them. They are framework. Yeah. They are, they are excellent. And they allow enough room for space and innovation. But they give the guidance that it's not just for education, but let's businesses and, and families sort of think about AI. Schools are desperate.
Every time I was running an MP AI session just before the summer holidays, and the feedback from that session was, this is all great and really interesting, but what do I do? What do I do? And they are desperate for really clear guidance and advice around what to do. And the sooner we can provide those parameters, here are some core principles and then exemplify that with case studies of how schools are choosing to interpret those principles for their unique contexts, because every school has different needs.
But we just need we need the sector is is grappling at the moment to try and find this. And, you know, if you happen to know a great school that's doing it, then you happen to learn about it. That's not equity for the system. We need a way to share the best practice effectively across the whole of the school estate.
Every time I was running an MP AI session just before the summer holidays, and the feedback from that session was, this is all great and really interesting, but what do I do? What do I do? And they are desperate for really clear guidance and advice around what to do. And the sooner we can provide those parameters, here are some core principles and then exemplify that with case studies of how schools are choosing to interpret those principles for their unique contexts, because every school has different needs.
But we just need we need the sector is is grappling at the moment to try and find this. And, you know, if you happen to know a great school that's doing it, then you happen to learn about it. That's not equity for the system. We need a way to share the best practice effectively across the whole of the school estate.
10:51
10:51
10:51
Speaker 2
-
Speaker 2
I mean, just to confirm, I don't think you did put words in my mouth. I think that was that was a fair reflection of our position. I think the point is around balance. You're kind of point. I think the kind of implied point you're making around, well, kind of suffocating innovation and constraining schools and teachers too much. I think that's a fair point and that's well made. I think my assessment would be that at the moment we don't have that balance right, particularly in a set of circumstances where there is, as we've been discussing, so much uncertainty in so many good questions to which there aren't at the moment particularly good answers.
10:52
10:52
10:52
Speaker 1
-
Speaker 1
If I could just add briefly, I think, yeah, it's about getting these balance. But I also think we've got to remember there's significant inequities in the in the schooling system that we're seeing and how schools can respond to policy and respond to technology. So I think what the DfE are doing in terms of digital infrastructure and so on is so important. We definitely not there, but finding ways to provide space for teachers to, to, to, to work out what is important in their context remains important.
But I agree, there's a huge amount of like duplication and churn within the system where people, where schools are individually trying to write down or evaluate evidence and all the rest of it.
But I agree, there's a huge amount of like duplication and churn within the system where people, where schools are individually trying to write down or evaluate evidence and all the rest of it.
10:53
Peter Swallow
-
-
Peter Swallow
So I wanted to come to, to disadvantage next actually. So a helpful segue there. The government's announced plans to expand AI tutoring tools to deliver personalised 1 to 1 support, I think particularly targeted at at students, you know, who need that additional support. How do you how do you see that working in practice? You know, what are the benefits, the opportunities and what are the risks there?
10:53
Rebecca Paul
-
-
Rebecca Paul
I'm happy to go first. Yeah, I've worked with a number of companies who are involved in some of those funded projects already, and seen under the engines of the technologies that they are using. We have to be so cautious about any AI driven systems that are directly facing students, learners. So there's a, you know, you can sort of play off. Right now, we have human tutors that are between many of the AI tools in these tutoring platforms and the students.
So the AI is advising the tutor on what move to make next with the learner. It's not directly advising the learner or directing the learner in any way. One of the things we know is that the difficulties that the underachieving children have are usually quite deep and highly interconnected with other aspects. So developing technologies, they tend to focus on the subject content. In other words, it's a version of the textbook with some fancy ins and outs and ups and downs.
But really many of the issues go broader than that. And it's conceptual problems and challenges that are underpinning the difficulties rather than quick fixes. So I think it's cautious on where the technology's at, but also cautious in whether that is the best use of time with those students, whether those students actually are the ones who most need human tutor interaction rather than agentic AI tutor interaction.
So the AI is advising the tutor on what move to make next with the learner. It's not directly advising the learner or directing the learner in any way. One of the things we know is that the difficulties that the underachieving children have are usually quite deep and highly interconnected with other aspects. So developing technologies, they tend to focus on the subject content. In other words, it's a version of the textbook with some fancy ins and outs and ups and downs.
But really many of the issues go broader than that. And it's conceptual problems and challenges that are underpinning the difficulties rather than quick fixes. So I think it's cautious on where the technology's at, but also cautious in whether that is the best use of time with those students, whether those students actually are the ones who most need human tutor interaction rather than agentic AI tutor interaction.
10:55
Peter Swallow
-
-
Peter Swallow
I want to come back to that human element because it's really important. But just very, very briefly, as a former teacher, I know one of the most challenging things to do in a classroom setting with sort of, you know, 30 different students in front of you, is that differentiation? I also have seen some AI tools that are very, very good at differentiation. It's one of the things that it seems to be best suited for. So is that one use case.
10:55
Rebecca Paul
-
-
Rebecca Paul
Where, I mean, the there's a diagnostic aspect, which there can be really much more quickly, much quicker than a human teacher in some ways, because they can connect up a bigger picture. But many of them assume that the learner has been on these platforms for hours and hours and hours and hours and hours to have sufficient data to do that fine tuning. And rarely does that happen. So there's there's an assumption, but one of those fundamental assumptions in how these AI's will work well isn't satisfied because you have to start somewhere.
10:56
Peter Swallow
-
-
Peter Swallow
And Anderson, earlier you you were very strong on the human aspect of teaching. And again, as a former teacher, I would agree. Is there a risk that if we overly rely on AI tutor tools to support the most disadvantaged students, they miss out on that really fundamental and valuable face to face time with teachers?
10:56
Speaker 3
-
Speaker 3
I think. So I think I think from everything we know about how learning happens and how relationships develop, we know there's a massive risk for that. And, you know, I wish there was this magic bullet where we could suddenly turn on great teaching automatically for disadvantaged pupils. I don't think it exists, and I think it is direct. It's sending our attention in the wrong way when we should be thinking about how do we really solve this issue of disadvantage, how do we get more high quality teachers into the classroom, rather than trying to use AI chatbots to solve that problem? And I think what often gets conflated is this idea of efficiency.
And the AI chatbots are more efficient. And there are schools that exist in America that are on that model of efficient learning. But if we ask ourselves, do we want children to have efficient learning? Like, is there an efficient poem or efficient song to sing? And teachers aren't about just about transmitting information. It is the relationship. It is showing people how to be kind and how to be courteous and how to wait.
And the idea that if we have this hyper personalized learning bubble the child sits in with a screen, when do they learn to tolerate others? When do they learn? It's okay to be a bit bored. It's okay not to have your every single need met every moment of the day. Because what kind of expectation are we creating for them for the rest of their lives? So I think we'll be doing them a great disservice if we switch them over to a screen AI tutor relationship, rather than having a teacher relationship.
And the AI chatbots are more efficient. And there are schools that exist in America that are on that model of efficient learning. But if we ask ourselves, do we want children to have efficient learning? Like, is there an efficient poem or efficient song to sing? And teachers aren't about just about transmitting information. It is the relationship. It is showing people how to be kind and how to be courteous and how to wait.
And the idea that if we have this hyper personalized learning bubble the child sits in with a screen, when do they learn to tolerate others? When do they learn? It's okay to be a bit bored. It's okay not to have your every single need met every moment of the day. Because what kind of expectation are we creating for them for the rest of their lives? So I think we'll be doing them a great disservice if we switch them over to a screen AI tutor relationship, rather than having a teacher relationship.
10:57
10:58
Speaker 1
-
Speaker 1
I would just say in our ethnographic work in secondary schools, we've definitely seen how these adaptive personalised systems do not really work that well, and they particularly don't work well for learners who are struggling academically or students with Send. And they find these very demotivating and kind of soulless experiences. And I think what's particularly important is that we're seeing that in schools that serve more economically disadvantaged communities, they're relying on these tools more often. So you're getting this kind of problematic effect where they're tending to rely on these options to to because they don't have enough teaching assistants or they have teacher shortages in some areas.
And so this is kind of compounding the problem. And we've got many ethnographic stories of students saying, I just want someone to sit with me while I go through this tool and teachers talking about how they're now sitting in a room with a number of students doing their literacy intervention on an adaptive tech, and they feel like, you know, there's this kind of third actor in the room where they're not engaging properly with their students. So I think there's some really important dynamics here that we have to think about.
So
And so this is kind of compounding the problem. And we've got many ethnographic stories of students saying, I just want someone to sit with me while I go through this tool and teachers talking about how they're now sitting in a room with a number of students doing their literacy intervention on an adaptive tech, and they feel like, you know, there's this kind of third actor in the room where they're not engaging properly with their students. So I think there's some really important dynamics here that we have to think about.
So
10:59
10:59
10:59
10:59
Speaker 3
-
Speaker 3
And I think if you look to sort of the corporate side and the big models that are out there, like the Khan Academy and the whole promise of what calming would produce and what Sal Khan said himself about the promise of this tech, and he's stepped back and said, actually, it can't deliver this personalized learning is not going to be what we thought it could be. And he's the most invested in saying the opposite. And he used to say that.
And I think we need to pay attention to that point.
And I think we need to pay attention to that point.
10:59
10:59
11:00
Helen Hayes
-
-
Helen Hayes
Thank you very much. It's been a really fascinating panel discussion this morning, and I'm grateful to all of you for taking the time to be with us. If there's anything that we didn't have the time to get into in as much detail as you'd like to additional points you want to make, please do write to us after the session and we'd welcome hearing from you again. But for now, thank you very much indeed. And if we could change over to the next panel as quickly as possible, that would appreciate it.
Thank
Thank
11:00
11:01
11:01
11:01
11:01
11:01
11:01
11:02
11:02
Speaker 6
-
Speaker 6
Thank you. There's an enormous range, actually. I think that's in many ways, the key thing for the committee's inquiry is that it ranges from deeply negative at one end, and we've got to be really, really wary and cautious and put in place mitigations or even bans and preventions through to massive opportunity. At the other end of the scale, where we've really got to be open to exploring the possibility for significantly improving quality.
In your previous body of evidence just taken, you've had quite a lot of talk about saving time and so on and so forth. I'm actually much more exercised about generally lifting the quality and performance of the education system, because actually, I think that's where the greatest gains are going to lie. Not in cutting costs, not in diminishing human, human, student or teacher student interaction, but actually enhancing that. And so I think the litmus test for me really is if it enhances human, the human teacher student relationship, we should absolutely be be looking at that if it seeks to displace, be very wary.
In your previous body of evidence just taken, you've had quite a lot of talk about saving time and so on and so forth. I'm actually much more exercised about generally lifting the quality and performance of the education system, because actually, I think that's where the greatest gains are going to lie. Not in cutting costs, not in diminishing human, human, student or teacher student interaction, but actually enhancing that. And so I think the litmus test for me really is if it enhances human, the human teacher student relationship, we should absolutely be be looking at that if it seeks to displace, be very wary.
11:03
11:03
Speaker 7
-
Speaker 7
Yeah, I agree mixed. And I would categorize it in slightly different ways in in how it's mixed. So in terms of the impact our Pearson School and college report showed, 64% of pupils are using AI for their homework, and 80% of them are using it every week. I think if you look at the overall assessment system, exams are
11:03
Speaker 1
-
Speaker 1
You look at the overall assessment system, exams are fairly similar to how they've been ten years ago. So there's kind of a mix of impact. And I think to Colin's point, there's also a mix of of good and bad. I think the key point there is AI itself isn't inherently good or bad if . pedagogically led. If it's. educationally led, if it's based on learning.
11:04
11:04
11:04
11:04
11:04
11:04
11:04
Rebecca Paul
-
-
Rebecca Paul
strategy in place. Thank you , Gillian. I think it's a bit different for the work we do, because obviously the impact on primary, I think, is lesser than in secondary. I, I look at it in the context of two things. One, in how we deliver our tests and then the other in how we develop our tests. And on the former, we are making very limited use of AI to enhance the development of tests.
So only in a very small way, we're making use of AI to do things like auto marking, but again, only in a very, very small way. But I think all the things that are faced by secondary. schools in relation to potential use of AI in. coursework and in.
So only in a very small way, we're making use of AI to do things like auto marking, but again, only in a very, very small way. But I think all the things that are faced by secondary. schools in relation to potential use of AI in. coursework and in.
11:05
11:05
11:05
Helen Hayes
-
-
Helen Hayes
You know, we're just not seeing that in the same way. Thank you. How should the assessment system evolve in response to the presence of AI in society? And I'd like to think both about the skills and capabilities that it should prioritise to enable young people to thrive in workplaces, many of which will be dominated by and enabled by AI in the future. But also, what are the skills and capabilities that it should prioritise to enable our young people in the future? To build a good society? In a society where AI is such a dominant and influential feature .
Of
Of
11:06
Speaker 1
-
Speaker 1
life. And I might start with Andrew Freeman this time, if that's okay. Okay. Thank you. I mean, I think the first. Point is really the challenge isn't preventing AI in education and in assessment. It's already happening, as I say, in terms of that. 64% of pupils using it every week, using it for their homework. And probably since that survey was done, that number has gone gone up.
I think the challenge is how do we ensure we've still got that valid. reliable and. meaningful and fair. assessment in in that AI enabled world that will be there. And related to that, those students obviously will be in that AI enabled world in the future. And in order to thrive, you need to be able to work with those AI tools. So I think the the core within assessment is really to move from output and really assessing.
output to looking at the process of learning. And I think one of the key skills is going to be learning to learn is going to be, you know, hugely important for their future, future success. Those higher order thinking skills. So things like problem solving, things like. reasoning and then application. And I would include within, within that communication. And I think the potential for AI within the assessment. system is actually to make the assessments more.
I think the challenge is how do we ensure we've still got that valid. reliable and. meaningful and fair. assessment in in that AI enabled world that will be there. And related to that, those students obviously will be in that AI enabled world in the future. And in order to thrive, you need to be able to work with those AI tools. So I think the the core within assessment is really to move from output and really assessing.
output to looking at the process of learning. And I think one of the key skills is going to be learning to learn is going to be, you know, hugely important for their future, future success. Those higher order thinking skills. So things like problem solving, things like. reasoning and then application. And I would include within, within that communication. And I think the potential for AI within the assessment. system is actually to make the assessments more.
11:07
11:07
Speaker 1
-
Speaker 1
To enable some of those things to be assessed. And as an example of that, we're doing a lot of work with AI personas internally. So we take a lot of data. We take a lot of market research about what a , say a science teacher might think, and then allow our teams to talk to that science teacher to get a sort of good insight based on all the collective knowledge that we have.
And you can kind of see that that is quite a small step to saying, could you do that in a, in a assessment world as well? So, for example, a, an AI persona that had dementia that you might use to actually .
And you can kind of see that that is quite a small step to saying, could you do that in a, in a assessment world as well? So, for example, a, an AI persona that had dementia that you might use to actually .
11:08
Speaker 2
-
Speaker 2
care. Yeah. I mean, actually picking up on, on what Andrew's saying, there is an interesting paradox, isn't there, that it's arguable that the skills that people are really going to need to be able to use AI effectively in their subsequent. post-education lives. And actually, is there a post-education life? That's a different question. Are actually the kind of skills in terms of higher thinking skills and so on and so forth, critical skills that Andrew is talking about there.
And if. And so I do think and let's call it out, we I personally believe we do have a problem about about coursework. There is an issue. around the degree to which it's possible for students to, you know. essentially to. replicate the. appearance of having thought about something where, in fact, they've used AI to do that. And in fact, we need to head in the other. direction and use AI to enable teachers really to, to track students developing those thinking and critical skills and so on and so forth.
And in part because when you know, as as students move on, the people who are now. deploying AI effectively, the people in my organisation who are really good at it are the people with those higher tier skills. So I think that's absolutely. And that leads me on to my second point, which is that digital literacy is absolutely fundamental here. And we have been really, really quite backward in moving in developing digital literacy in our schools and colleges.
And the problem there is partly that there is a rather, in my view, rather rather slack view that all young people are digital natives. And that is utter nonsense. I mean, our own. research illustrates that, you know, students, three out of four students are extremely comfortable with TikTok and Snapchat. Actually, probably only about half will say that they're comfortable even using word. Probably the most basic thing they're going to use out there in the workplace.
They're not comfortable with PowerPoint. They're not comfortable with Excel. And so I think so I think there is a big job to be done and simply hoping that it'll just.
And if. And so I do think and let's call it out, we I personally believe we do have a problem about about coursework. There is an issue. around the degree to which it's possible for students to, you know. essentially to. replicate the. appearance of having thought about something where, in fact, they've used AI to do that. And in fact, we need to head in the other. direction and use AI to enable teachers really to, to track students developing those thinking and critical skills and so on and so forth.
And in part because when you know, as as students move on, the people who are now. deploying AI effectively, the people in my organisation who are really good at it are the people with those higher tier skills. So I think that's absolutely. And that leads me on to my second point, which is that digital literacy is absolutely fundamental here. And we have been really, really quite backward in moving in developing digital literacy in our schools and colleges.
And the problem there is partly that there is a rather, in my view, rather rather slack view that all young people are digital natives. And that is utter nonsense. I mean, our own. research illustrates that, you know, students, three out of four students are extremely comfortable with TikTok and Snapchat. Actually, probably only about half will say that they're comfortable even using word. Probably the most basic thing they're going to use out there in the workplace.
They're not comfortable with PowerPoint. They're not comfortable with Excel. And so I think so I think there is a big job to be done and simply hoping that it'll just.
11:10
11:10
Speaker 2
-
Speaker 2
In the process of teaching other subjects, we we are sceptical about that, frankly, as as a Q a and then further down the line, I think that some of the things that your previous witnesses were talking about, I think there are really serious. opportunities in terms of. ascertaining a much more rounded picture of where a student is at in at any given point in time. using artificial intelligence tools, for example, for. diagnostics and so on and so forth.
And therefore providing a body of. data which enables enables us properly to respond to each .
And therefore providing a body of. data which enables enables us properly to respond to each .
11:11
Rebecca Paul
-
-
Rebecca Paul
So those three points. Thank you. So again, I think the context is very different for primary. I mean, the obviously stress tests, we don't test computing. We don't test it. We do the basic skills. But the research has said that, you know, the importance of alongside developing digital literacy, you need to still have those basic skills, those basic knowledge. So I think we support it by ensuring that pupils have basic maths, have basic have basic literacy.
The other thing I think that I think is of, of relevance to the work that that Sta does is ensuring we don't do anything that extends the digital divide. So in all the work we've done in relation to more digital.
The other thing I think that I think is of, of relevance to the work that that Sta does is ensuring we don't do anything that extends the digital divide. So in all the work we've done in relation to more digital.
11:11
11:11
11:12
11:12
Rebecca Paul
-
-
Rebecca Paul
That there are options to do things off screen, but also. familiarisation work so that all pupils come at it from the same level. I think the third point I'd make in relation to the the potential risks of AI are how we. What is the risk of having data that is somehow exposed to AI, and that pupil data is somehow put into AI systems. And that's something that obviously, all of us who have databases that have pupils with with their results on take really, really seriously.
So I think that's one of the risks of having an AI model that has access to our pupil data.
So I think that's one of the risks of having an AI model that has access to our pupil data.
11:12
Helen Hayes
-
-
Helen Hayes
Seems to me a risk and something we need to be very, very careful of. Thank you very much. And then just finally for for me, the government has made a big announcement over the summer about technical education in schools and wanting to see an expansion of that. And I just wondered if you have any thoughts, quite briefly, on what implications the growing use of AI might have for . That
11:13
Speaker 2
-
Speaker 2
desire to expand technical education in schools. So a, this this is very welcome news. You know, we have long believed as an organisation that we we need significantly to improve in this area. And schools are begging for it, frankly. But there is there is the other interesting thing here, I think, is that the use of technology in general. digital tools, if you like, is actually has a longer and better and, and more deeply a.
customization in. vocational teaching and learning than it has so far in more academic, let's say, teaching and learning. So I think this is one of the areas where there is actually real. opportunity because, for example, the opportunities for. simulation of practical. activities is very, very high. These things are getting steeply cheaper to be able to do. They are fundamentally not about replacing teachers with machines.
They are they are using machines to enable the teacher to stand. 1 to 1 side and educate and assess. So it really, really meets my test, if you like. Does it. enhance the ability, the the quality of the human student interaction? So I think there's lots and lots of opportunity there. I do think having said that, that we also need to be very, very clear about the risk that people try, that we end up measuring a student's.
ability to use artificial. intelligence and not. measuring their ability to understand the.
customization in. vocational teaching and learning than it has so far in more academic, let's say, teaching and learning. So I think this is one of the areas where there is actually real. opportunity because, for example, the opportunities for. simulation of practical. activities is very, very high. These things are getting steeply cheaper to be able to do. They are fundamentally not about replacing teachers with machines.
They are they are using machines to enable the teacher to stand. 1 to 1 side and educate and assess. So it really, really meets my test, if you like. Does it. enhance the ability, the the quality of the human student interaction? So I think there's lots and lots of opportunity there. I do think having said that, that we also need to be very, very clear about the risk that people try, that we end up measuring a student's.
ability to use artificial. intelligence and not. measuring their ability to understand the.
11:14
Speaker 1
-
Speaker 1
Material that they're grappling with. Yeah, I mean, I'd agree with with a lot of what Colin has said in that I think there is tremendous. opportunity in that space. I think the key. points with the assessment of vocational. education are that it is properly practical and what that means. And I think the thing about vocational education is, is as with AI, it's not it's not it's so varied.
So it means different things in different places.
So it means different things in different places.
11:15
11:15
Speaker 1
-
Speaker 1
A, a may be a useful additional tool within that assessment process. And sometimes it may, it may not be, but it needs to be practical and relevant and authentic. It needs checkpoints to, to the point , again, around it's not just about the output. And we have to move away from someone producing something very shiny at the end that isn't what's being assessed. It needs to be assessment right the way through, partly as a as a safeguard, a sort of.
triangulation that, you know, the work has been done and it's of that quality, right, right the way through, but also the, the, the, the point about learning to learn being really fundamentally important for their futures. And then thirdly, there's probably an element of supervision as well for some elements of it as, as part of that, as part of that safeguard. And then just finally, I'll just say one, one other thing that we're looking at at the moment is an AI viva tool.
Viva tool. So that there's the. opportunity a to sort of push on those oracy skills, but B, to have another. checkpoint of, you know, how is that student actually able to talk about their work as well . As
triangulation that, you know, the work has been done and it's of that quality, right, right the way through, but also the, the, the, the point about learning to learn being really fundamentally important for their futures. And then thirdly, there's probably an element of supervision as well for some elements of it as, as part of that, as part of that safeguard. And then just finally, I'll just say one, one other thing that we're looking at at the moment is an AI viva tool.
Viva tool. So that there's the. opportunity a to sort of push on those oracy skills, but B, to have another. checkpoint of, you know, how is that student actually able to talk about their work as well . As
11:16
11:16
11:16
11:16
11:16
11:16
11:16
Speaker 1
-
Speaker 1
Yeah, I can start with that one. I mean, at Pearson we've we've been using again, AI isn't all one thing I would say on screen again, actually isn't all one thing. We've been doing the Pearson test of English for 15 years now using machine learning, a kind of offshoot of AI. So not generative AI, but machine learning AI. And we assess a million or so learners a year using using that on screen methodology.
So we've got a lot of evidence that that is really robust, that it works, it's accepted by governments, it's accepted by institutions around the world. And we've also got a lot of evidence from our international GCSE. So we do thousands of international. GCSEs on screen. And I think to me, the biggest risk in assessment is, is really that we we lock it down and we're left behind on the assessment side from what is happening on the teaching and learning side, where it is very digital.
And for me, the three big. opportunities are, are two of them are around authenticity. So
So we've got a lot of evidence that that is really robust, that it works, it's accepted by governments, it's accepted by institutions around the world. And we've also got a lot of evidence from our international GCSE. So we do thousands of international. GCSEs on screen. And I think to me, the biggest risk in assessment is, is really that we we lock it down and we're left behind on the assessment side from what is happening on the teaching and learning side, where it is very digital.
And for me, the three big. opportunities are, are two of them are around authenticity. So
11:18
11:18
Speaker 1
-
Speaker 1
To what is that student doing every day in terms of their teaching and learning, but also outside of that. authenticity to what they end up doing after they leave the education sector or go to the next stage, whether it's university or work. Again, digital and AI is really increasingly implicit in so many roles that are being done or that that further study and then on, on the accessibility point. Yeah, we've seen like real sort of benefits on the, on the.
accessibility side, when you compare it with, for.
accessibility side, when you compare it with, for.
11:18
11:18
11:18
11:18
11:19
11:19
Speaker 2
-
Speaker 2
I think one you you touched on, which is that it's potentially fairer. We can actually, for example, deal with most special, not all, but most special needs issues can be dealt with better in in on screen assessment. It's more resilient. I really worry about the fact that people have already. forgotten that. I was asked in.
11:19
11:19
Speaker 2
-
Speaker 2
I deliver Maths and English GCSEs online in 2021? And the answer was no, because nobody's ever asked us to do that, and we would have been in a much more resilient place during the pandemic if we'd already got experience in delivering digital exams. Security is probably going to be better. But the really big appeal, apart from I support everything that Andrew said, I'm trying to add to it, is potentially more rounded assessment.
We will be able to assess different things in different ways, and I'm not I don't think we should replace. We absolutely must not. replace all of our. examinations with digital or on screen examination. That's a really terrible place to go. Firstly, as you can see, I'm very fond of pen and paper. Pen and paper, we know has lots of advantages. It actually has cognitive advantages. It's one of the.
mediums with media that we should continue to use, but there are certain things that we can't easily examine with pen and paper. You know, think about the business of. interacting with maps in. geography or, or, you know, actually. replicating scenarios in physics and so on and so forth. There are all sorts of things that we, we could potentially do in due course using online. So I think the key thing is to match the media to the thing that, that we actually want to assess and not get obsessed with this idea that it must all be pen and paper or it must all go digital.
That's not the right, not the right way to think about it. So I think there are very significant gains to be had. We are very strongly in favour of taking taking.
We will be able to assess different things in different ways, and I'm not I don't think we should replace. We absolutely must not. replace all of our. examinations with digital or on screen examination. That's a really terrible place to go. Firstly, as you can see, I'm very fond of pen and paper. Pen and paper, we know has lots of advantages. It actually has cognitive advantages. It's one of the.
mediums with media that we should continue to use, but there are certain things that we can't easily examine with pen and paper. You know, think about the business of. interacting with maps in. geography or, or, you know, actually. replicating scenarios in physics and so on and so forth. There are all sorts of things that we, we could potentially do in due course using online. So I think the key thing is to match the media to the thing that, that we actually want to assess and not get obsessed with this idea that it must all be pen and paper or it must all go digital.
That's not the right, not the right way to think about it. So I think there are very significant gains to be had. We are very strongly in favour of taking taking.
11:20
11:21
Speaker 2
-
Speaker 2
In this direction. We are not in favour of. jumping hell for leather for, you know, sort of total system on screen. examination approach. But I do actually, I am concerned that we are at risk of. moving too slowly and, and at too small a scale. We need to do things at a at a slightly bigger scale than is currently being discussed. And we need to do a slightly wider range of subjects for the reasons that Andrew has just given in each subject domain.
11:21
11:21
Speaker 2
-
Speaker 2
It's going to be different. And until we get the. opportunity to, to, to start doing this, not with hundreds of thousands. AKA examines. 800,000 GCSE English language students a year. I'm not proposing to go there. I do think that we should look at tens of thousands of students, not. 2 or 3000 students, and we should look at 4 or 5 subjects, not not just two subjects, but having made that one sort of relatively small.
11:22
11:22
11:22
Rebecca Paul
-
-
Rebecca Paul
Moderation approach is definitely the right approach. So we already have two on screen assessments that are used with around. 600,000 pupils a year. And so we've seen quite a lot of benefits. The reasons we introduced them as digital, fully digital assessments. One was accessibility. So the contrast is with our key stage two assessments, the SATs, we send out millions of pieces of paper, I mean hundreds of millions of pieces of paper around the country.
And then they all have to be collected in. There's a vast amount of workload for the teachers and for the admin staff in schools in relation to doing that sort of support for tests, for the tests that we have on screen. There's none of that. There's environmental impact. Obviously, there's an environmental impact of digital. I'm not saying there isn't, but there is an environmental benefit through not sending out a lot of paper.
But I'd say the one that's the that you touched on that is the most important one is. accessibility for pupils with send. So again, by contrast with our paper based key stage two.
And then they all have to be collected in. There's a vast amount of workload for the teachers and for the admin staff in schools in relation to doing that sort of support for tests, for the tests that we have on screen. There's none of that. There's environmental impact. Obviously, there's an environmental impact of digital. I'm not saying there isn't, but there is an environmental benefit through not sending out a lot of paper.
But I'd say the one that's the that you touched on that is the most important one is. accessibility for pupils with send. So again, by contrast with our paper based key stage two.
11:23
11:23
11:23
11:23
Rebecca Paul
-
-
Rebecca Paul
Or that where the they need a more. simplified image where we have those, those. digitally that can all be done through the system itself. There's not a teachers having to photocopy onto a different colour paper, that kind of thing, which happens a lot with the key stage two stats. So I think we are seeing through those tests that we're already doing benefits. And then on top of that, I would say the the ability to collect data pretty much instantaneously.
So both of those tests have an element of auto marking only where there's a single unequivocal answer. But where that happens. And then with the reception baseline, the rest of it is inputted by a teacher on screen. The data collection is instantaneous. There isn't the kind of. waiting weeks, well, marking takes place, and we can also get data back on how the system has been interacted with so that we can improve it year on year.
So I think those are quite. significant benefits that we've seen through our two statutory tests that . We've
So both of those tests have an element of auto marking only where there's a single unequivocal answer. But where that happens. And then with the reception baseline, the rest of it is inputted by a teacher on screen. The data collection is instantaneous. There isn't the kind of. waiting weeks, well, marking takes place, and we can also get data back on how the system has been interacted with so that we can improve it year on year.
So I think those are quite. significant benefits that we've seen through our two statutory tests that . We've
11:24
11:24
Rebecca Paul
-
-
Rebecca Paul
extra infrastructure. equipment and capacity in the schools? I mean, we're not we have no specific plans to do at the moment. We have done various bits of research over time about extending how we use digital, for example, with more auto marking. But in terms of something like the key stage two test, which is very high stakes, we would have to do, we would take it very cautiously. We would have to do a lot of.
comparability tests to make sure that like for like was, was, was being measured. We have a standard for the key stage two test, which is set back to. 2016 standard. We'd have to test all of that. We wouldn't want to be doing the putting the test behind glass, which is one option. We'd want to develop a test that was. purposefully for on screen, and we don't have the plans to do any of that yet.
It's it's a possibility for the future, but I think we'd have to have a lot of assurance that it was as good a test. And that's also you've mentioned. infrastructure with our existing on screen tests. They don't require a class of 30 pupils or two classes of 30 people sitting down at the same time, which for a primary school implies quite a lot of technical infrastructure, that they probably don't have connectivity that might be beyond what they have.
So yes, I don't think we would at the moment feel like . Confident
comparability tests to make sure that like for like was, was, was being measured. We have a standard for the key stage two test, which is set back to. 2016 standard. We'd have to test all of that. We wouldn't want to be doing the putting the test behind glass, which is one option. We'd want to develop a test that was. purposefully for on screen, and we don't have the plans to do any of that yet.
It's it's a possibility for the future, but I think we'd have to have a lot of assurance that it was as good a test. And that's also you've mentioned. infrastructure with our existing on screen tests. They don't require a class of 30 pupils or two classes of 30 people sitting down at the same time, which for a primary school implies quite a lot of technical infrastructure, that they probably don't have connectivity that might be beyond what they have.
So yes, I don't think we would at the moment feel like . Confident
11:25
11:25
11:25
Speaker 2
-
Speaker 2
digital is, I would say, cautious. Same question to Colin. Thank you. Subject to regulatory approval, we obviously we obviously in constant and I have to say, friendly discussion with Ofqual around this topic. So anything I say in this area is not is not is not an argument with Ofqual. We are we're in a good place about discussion about this as Aqa. We are absolutely ready to start doing this in a range of areas and want to to get on with it, but one of the reasons that we want to do, you know, do it at moderate scale to begin with, is precisely the point that you talk about.
Actually, schools by and large do have the, the. machinery and so on. And so they have the connectivity because we'll be downloading assessments and then uploading them afterwards, they won't be online during the exam. The the issue really actually. overwhelmingly in, in schools where they do have problems is space. And so, you know, having. 250 students in an exam hall all doing one big.
examination is, is definitely a problem tomorrow morning. But for example, doing a minority language. GCSE or a relatively small A-level like philosophy or something like that is not particularly problematic. And, and we can and we can take those. incremental steps, if you like, and demonstrate, because the key thing we've got to do is carry public confidence with us around this. And so just going gung ho is not going to be a good idea.
The the, the interesting thing when we talk to.
Actually, schools by and large do have the, the. machinery and so on. And so they have the connectivity because we'll be downloading assessments and then uploading them afterwards, they won't be online during the exam. The the issue really actually. overwhelmingly in, in schools where they do have problems is space. And so, you know, having. 250 students in an exam hall all doing one big.
examination is, is definitely a problem tomorrow morning. But for example, doing a minority language. GCSE or a relatively small A-level like philosophy or something like that is not particularly problematic. And, and we can and we can take those. incremental steps, if you like, and demonstrate, because the key thing we've got to do is carry public confidence with us around this. And so just going gung ho is not going to be a good idea.
The the, the interesting thing when we talk to.
11:27
11:27
Speaker 2
-
Speaker 2
You know, school leaders, mat leaders and so on and so forth, they, they have on the one hand a sort of, you know, wariness, a little bit of reluctance. Have we got all the kit? Are we ready for this? And so on and so forth. And then you say to them, but do you think we should go there? And they . Say.
11:27
11:28
11:28
Speaker 1
-
Speaker 1
Our experience on the international GCSE. side has been very much as Colin said, it's kind of incremental. So schools take time to kind of think through whether they want to be part of a programme to go on screen. They kind of move in a in a slow but considered manner. So it's not a kind of from zero to an enormous number in a very short period of time. So our.
11:28
11:28
11:28
11:28
11:29
11:29
Helen Hayes
-
-
Helen Hayes
Especially those with sound or disabilities? I'm just sorry to interject. We are at the we've got about 15 minutes left. I need to end this panel promptly at quarter two so that we can get in our final witness. So I do need to ask you to be extremely brief, because we've got quite a few more topics to cover,
11:29
Speaker 2
-
Speaker 2
If that if if that's okay that we want to ask you about. Thank you. I think the big one is, is this question about, you know, the so-called digital divide. And for me, that is actually the one of the arguments in favour of moving exams digital is that actually it will then mean that schools need to act on that. So but in terms of special needs and so on, I think there is a there is .
11:29
11:29
11:30
11:30
11:30
11:30
11:30
11:30
11:30
11:30
11:31
11:31
11:31
11:31
11:31
11:31
11:31
Speaker 2
-
Speaker 2
That Aqa delivers and about. 10.5 million scripts. So it gives you it gives you some notion of, of what a small proportion and of the total. malpractice issues raised with us. It's only. 2.4% of the total. So and but interestingly, so it's, you know, it's. relatively minor at this stage. I think the. concern is much more the degree to which. teachers looking at coursework, for example, in A-level English literature or A-level.
history are are now have an extra burden on themselves to try and work out. Has a has my student used AI here? And it's a real. I think it's a genuine struggle for them. And what in fact is tending to happen is moderators are actually slightly over reporting their concerns about that. So which I kind of understand, that seems logical and natural. My personal view is that we, we will probably need to think during this process of curriculum and assessment reform.
All of this is being reviewed at the moment, as you as you very well know. And I think we're going to need to look at that question, not because we think there's a lot of cheating going on. We are very, very good. We do trouble somersaults backwards to identify cheating. We we kind of know what we're doing. We've got long history in doing that, that I think the problem is more about the burden on teachers and, and the difficulties they have and the worry that they experience in trying to.
history are are now have an extra burden on themselves to try and work out. Has a has my student used AI here? And it's a real. I think it's a genuine struggle for them. And what in fact is tending to happen is moderators are actually slightly over reporting their concerns about that. So which I kind of understand, that seems logical and natural. My personal view is that we, we will probably need to think during this process of curriculum and assessment reform.
All of this is being reviewed at the moment, as you as you very well know. And I think we're going to need to look at that question, not because we think there's a lot of cheating going on. We are very, very good. We do trouble somersaults backwards to identify cheating. We we kind of know what we're doing. We've got long history in doing that, that I think the problem is more about the burden on teachers and, and the difficulties they have and the worry that they experience in trying to.
11:32
11:32
Speaker 2
-
Speaker 2
Identify that. And as I said earlier, the concern that we might end up actually examining a student's ability to play with AI and not their ability to actually show that they've really thought about understood and, you know, a Shakespeare play or whatever, whatever it may be. So that's going to need there is a simple solution to that, which is just to switch it all into a closed exam room. There is a much more. exciting answer to that, which I think Andrew was touching on, which is the.
potential for actually having digital.
potential for actually having digital.
11:33
11:33
11:33
11:33
Rebecca Paul
-
-
Rebecca Paul
Upstream, if you like. Yeah, I think the others want to comment. So we have a slightly different context in primary because we have some teacher assessed work on writing, which is part of the school accountability system, part of the national headline statistics. We haven't previously seen a problem with this, and I would have imagined that we wouldn't, because teachers through that process assess a portfolio of writing so they will see a whole range of writing, not a single piece of high stakes coursework.
And also because the teacher is the one doing the assessing so they will understand what their pupil can and can't do, and hopefully be able to identify where something looks quite not quite right. For the first time this year, though, we've had reports from local authority moderators who moderate the teacher. judgements saying that they've started to detect the use of AI either in the writing or in the editing of that work. We haven't had any. maladministration reports yet and we haven't seen any evidence, but I think it's something we've built into our guidance now, clarity, where it probably wasn't clear before, that independent writing cannot.
include anything that's using AI. But of course, the other thing that we have to allow for is that every year we get .
And also because the teacher is the one doing the assessing so they will understand what their pupil can and can't do, and hopefully be able to identify where something looks quite not quite right. For the first time this year, though, we've had reports from local authority moderators who moderate the teacher. judgements saying that they've started to detect the use of AI either in the writing or in the editing of that work. We haven't had any. maladministration reports yet and we haven't seen any evidence, but I think it's something we've built into our guidance now, clarity, where it probably wasn't clear before, that independent writing cannot.
include anything that's using AI. But of course, the other thing that we have to allow for is that every year we get .
11:35
11:35
11:35
11:35
Speaker 1
-
Speaker 1
I wouldn't be surprised if we saw some of that. And yes, similarly, our numbers are very, very low of a kind of similar order in terms of AI sort of. reported abuse to, to to date, we've got checks and balances in place that we work with Jake, with Ofqual, to basically ensure that the centre is checking both a human level right the way through the.
11:35
11:35
11:35
11:35
Speaker 1
-
Speaker 1
But kind of taking a step back and looking at an assessment. holistically and going back to that stat of. 64% of children using an LLM for for their homework. clearly AI is in the mix in, in assessment, in totality. And one of the things we asked about in the Pearson School report was your feeling about whether you've prepared your students for an AI in in enabled world.
And only 8% of teachers said that they , they feel confident that they're doing that. And I think there's a number of things. associated with that. We've talked about infrastructure a little bit, but professional development is really, really critical here. And I think that professional development . With
And only 8% of teachers said that they , they feel confident that they're doing that. And I think there's a number of things. associated with that. We've talked about infrastructure a little bit, but professional development is really, really critical here. And I think that professional development . With
11:36
11:36
11:36
11:36
11:37
11:37
11:37
11:37
Speaker 2
-
Speaker 2
What we are doing, and we have piloted now for two years and we're in effect very nearly ready to go, is having AI mark the markers. So it's sitting on the human. shoulder watching the human. And the evidence is crystal clear that we can massively jack up the quality of marking in, you know, because we're watching them. The machine is watching them the real time. And then alerting another sort of super examiner human.
This is, you know, there is there is a consistency question here that's saving a huge amount of time, lifting the quality of the human marking. And I think that's where the game is at at the moment. As far as high stakes. examinations are concerned, it it must be clear to you, of course, though, that while we're doing that, guess what the machine is, is getting to understand a huge amount about how marking is done and happens.
And so there will be, of course, a second phase to all of this. Interestingly, when I ask. multi-academy trust.
This is, you know, there is there is a consistency question here that's saving a huge amount of time, lifting the quality of the human marking. And I think that's where the game is at at the moment. As far as high stakes. examinations are concerned, it it must be clear to you, of course, though, that while we're doing that, guess what the machine is, is getting to understand a huge amount about how marking is done and happens.
And so there will be, of course, a second phase to all of this. Interestingly, when I ask. multi-academy trust.
11:38
11:38
11:38
11:38
Speaker 2
-
Speaker 2
You know, how do you feel about about this, the introduction of AI into marking and their response is, you know, a huge welcome. They, they've we've shown them we've talked to them about what we're doing. And they can see immense attraction because part of the issue at the moment is that, you know, we've got. 38,000 teachers marking exam papers every summer. AI what we're hoping AI will do, in fact, the evidence is that it will do is it will enable us to be far more consistent .
Across that
Across that
11:38
11:38
11:38
Speaker 1
-
Speaker 1
markers. Probably. Can I just just very quickly, just on on the Pearson test of English, because we've been doing it for such a long time. And, you know, it's a million, a million assessments happening every every year. We've got a lot of research on, on that, which we'd be very happy to share. But but.
11:39
11:39
11:39
11:39
11:39
Rebecca Paul
-
-
Rebecca Paul
Yeah. Thank you. Anything to add, Gillian? I only to say, I think human interaction with the AI is the crucial thing. And where we've used AI to train them, train the moderators of the markers that has needed a lot of human interaction because AI makes things up sometimes. And so not only does it amplify biases, I think it also invents stuff. And so I think the crucial thing in all of .
It is
It is
11:39
11:39
11:39
11:39
11:39
11:39
11:40
11:40
11:40
11:40
Speaker 1
-
Speaker 1
Covers leadership and teaching practice, and covers curriculum and assessments. We would see AI not as something that you you have as a discrete subject. It's something that has to be embedded across the. curriculum in, in all subjects because it's applicable in, in different ways. We're very proud to be part of the government AI skills BES programme. So I've got experience on that front. And also looking at a programme with Doctor Fiona Aubrey Smith in terms of AI literacy and how to best embed that.
So . Again,
So . Again,
11:41
Speaker 2
-
Speaker 2
Information on that. Yes please. Given time. Sure. Yeah, that'd be helpful. Thank you. I'd be wary of. embedding AI in teaching and learning, but I would not be I would be strongly in favour of. enhancing the teaching of digital literacy and not just. hoping because it's. overoptimistic to expect it to simply arise out of subject teaching. We need to do something specific and. distinct in terms of enhancing digital literacy.
We're also in favour of looking at the potential for AI or future technology. qualifications that could actually assist with all of this. I think actually the gains in the risks in teaching and learning are quite high. I think the gains in assessment are already demonstrated to be pretty positive. I haven't mentioned, for example, you know, we, we, we at a subsidiary of Acca delivers all of the adaptive. testing in key stage three in Wales.
Huge teacher wariness about it until it's actually done. And then they realise that it's just providing with them the most gorgeous body of data that they've ever had. And it's and it's completely wonderful. And then they embrace it and they're happy to go forward with it. So I think part of the issue here is we need to not to be too shy about actually getting out there and making some stuff happen.
Some of it will work, some of it won't. We don't take high risks with high stakes.
We're also in favour of looking at the potential for AI or future technology. qualifications that could actually assist with all of this. I think actually the gains in the risks in teaching and learning are quite high. I think the gains in assessment are already demonstrated to be pretty positive. I haven't mentioned, for example, you know, we, we, we at a subsidiary of Acca delivers all of the adaptive. testing in key stage three in Wales.
Huge teacher wariness about it until it's actually done. And then they realise that it's just providing with them the most gorgeous body of data that they've ever had. And it's and it's completely wonderful. And then they embrace it and they're happy to go forward with it. So I think part of the issue here is we need to not to be too shy about actually getting out there and making some stuff happen.
Some of it will work, some of it won't. We don't take high risks with high stakes.
11:42
11:42
11:42
11:42
Speaker 2
-
Speaker 2
We also have a platform that looks at key stage four mathematics diagnostics. It's called GCSE stride instead free to schools. And that is using both. adaptive and and machine learning. systems to. enable the. The machine can do things that the teacher finds really difficult as a maths teacher, which is really to. understand when they arrive at the beginning of year ten, where are there gaps in.
11:43
11:43
Speaker 2
-
Speaker 2
know, in their. conceptual understanding of mathematics? That's the biggest single problem. All of our diagnostic testing in other parts of the world, we're also currently doing it in Scotland, shows us that actually something like 1 in 4 students are no further forward in mathematics in year 11 than they were in year seven. Identifying that through just human interaction is very difficult. Using these kind of.
diagnostic technologies is telling us stuff that should be having a major policy impact .
diagnostic technologies is telling us stuff that should be having a major policy impact .
11:43
11:43
11:44
11:44
11:44
11:44
11:44
Rebecca Paul
-
-
Rebecca Paul
keep up with the speed of AI? And I'll go along the line. So, Gillian, I mean, again, I think the question is not relevant to the work we do in primary. But I think where we have used AI, I think the key thing that we that the regulatory system will need to be sure of is that there is fair assessment of.
11:44
11:44
Rebecca Paul
-
-
Rebecca Paul
11:44
11:44
Speaker 2
-
Speaker 2
how challenging that is for Ofqual to oversee. As Gillian says, you're about to hear from the chief regulator. I'm pretty confident that he and I would would not disagree about this. We think that there are, as I said earlier, issues with regard to. coursework and and how we manage that going forward. So I think we need to pay close attention to that as we're developing the new assessment regime over the next two, 3 or 4 years, that's going to be a primary area.
But I think the answer is, I'm afraid, is actually we just need to enhance digital literacy amongst teachers and. students more broadly. And by the way, the ethical background to all . Of that,
But I think the answer is, I'm afraid, is actually we just need to enhance digital literacy amongst teachers and. students more broadly. And by the way, the ethical background to all . Of that,
11:45
11:45
11:45
11:45
Speaker 1
-
Speaker 1
We're working very closely with Ofqual, and lots has already been said on that. I think on the more. general assessment within schools, there's there's quite a lot of different requirements and. frameworks and all of that being brought together, which is which is the plan will be very helpful. So there's things around data processing, child appropriate design, cognitive safety, AI safety, there's keeping children safe in education, there's GDPR, there's.
11:46
11:46
11:46
11:46
Helen Hayes
-
-
Helen Hayes
is in place is really welcome. Thank you very much. Thank you very much indeed. Apologies that we were a bit rushed. If there's anything that we you would like to tell us about in more detail, please do write to the committee after the session. We very much welcome that input, but thank you very much for taking the time to be with us today.
11:46
11:47
11:47
11:47
11:47
11:47
Speaker 6
-
Speaker 6
AI and edtech overall on the assessment system that we have? Yes, certainly. I think before getting to a detailed answer to that question, it's interesting to step back for a moment and just. reflect on what the purpose of education itself is, and therefore, what's the purpose of. educational assessment is in education, which is an enormously. expensive and also.
11:48
11:48
11:48
11:48
11:48
11:48
11:48
11:48
11:48
Speaker 6
-
Speaker 6
Because if we don't do that, then we're not able to make informed decisions about which young people should go to which destination. And neither are we able to. understand which parts of our system are working well and which are not. And to achieve all of that, then the next step in the logic chain is to say that educational assessment. must tell the truth. It must have authenticity. Now, across our suite of.
qualifications and assessments, we have many which are. undertaken in supervised.
qualifications and assessments, we have many which are. undertaken in supervised.
11:49
11:49
Speaker 6
-
Speaker 6
Which do have a high level of authenticity. We have we have, for example, written tests. We have the fabrication of a physical artifact that can be evaluated by an expert. We have spoken tests that can be. recorded high levels of authenticity. In a smaller number of cases, we have. assessments which are essentially. extended writing.
11:49
11:49
11:49
11:49
Speaker 6
-
Speaker 6
My view is that we are exposed to. risk in the shape of AI being. marshaled to support students to undertake those assessments. And I think that. compromises the degree to which those. assessments really tell us the truth. So I think we can track this all the way back to the importance of. education as an.
11:50
11:50
11:50
11:50
11:51
11:51
Speaker 6
-
Speaker 6
We look, for example, at the undoubted benefits of coursework on non examined. assessment for motivating students over a period of time for for developing their research skills and so on. And I think what we've got to understand now when we're thinking about that is that that comes with a very significant.
11:51
11:51
11:51
11:51
11:51
11:51
11:51
11:52
11:52
Speaker 6
-
Speaker 6
A confusion between on screen and. online assessments. So I just want to say up front that I regard those as two very different things. Good on screen assessment. utilises a secure portal. That means that students can't. access the wider. internet and can't. communicate with other students or other.
11:52
11:52
11:52
11:52
11:52
11:52
11:52
11:52
11:52
11:53
Speaker 6
-
Speaker 6
Using an on screen. format than is possible on paper. I think it's also the case that some special. educational needs, and I say some advisedly, because it's a very broad spectrum, some special special educational needs can be more easily catered for on screen than is possible on paper. So there's a couple of. benefits in terms of risks. The first one is scale of. implementation and.
complexity of implementation. We have a large number of schools and also. colleges where qualifications qualifications that we regulate are taken up and down the country. And we have uneven access to high quality, high quality digital. infrastructure and devices. There are also possible social and.
complexity of implementation. We have a large number of schools and also. colleges where qualifications qualifications that we regulate are taken up and down the country. And we have uneven access to high quality, high quality digital. infrastructure and devices. There are also possible social and.
11:53
11:53
11:54
11:54
Speaker 6
-
Speaker 6
Systems and. applications over a protracted period of time than others. So we have to bear that in mind as well. And I think there's also something to note about the. signalling that the widespread use of on screen assessment in terminal assessment gives to schools and the system, that the signal being that we expect learning and assessment to be on screen, and therefore that has a washback effect into schools and increases the.
expectation that teaching and learning should be on screen. And I would have to say that that is a contested.
expectation that teaching and learning should be on screen. And I would have to say that that is a contested.
11:54
11:54
11:54
11:54
11:55
11:55
Mark Sewards
-
-
Mark Sewards
need that a student has. So that already exists. Just very briefly on your point about Send, you were talking about some of the advantages. I can imagine some of the specific. scenarios where somebody has a need that can be met by an on screen assessment. And does any of the evidence, has any of your . Consultation
11:55
11:55
Speaker 6
-
Speaker 6
and said that there are. disadvantages to certain send pupils? Yes, absolutely. So Send is a very broad spectrum. Not all Send needs are the same by any stretch of the. imagination at all. I mean some of the simple and obvious things that an on screen format can achieve that paper can't are things like different fonts, colours, different text size, reduction of the amount of text on a screen and so on, which is very useful for some students with some needs.
But other send students actually find the digital interaction and the navigation of a keyboard more. difficult than a pen and paper format. So it's not an . Even
But other send students actually find the digital interaction and the navigation of a keyboard more. difficult than a pen and paper format. So it's not an . Even
11:55
11:55
11:55
11:56
Speaker 6
-
Speaker 6
And to. address AI enabled malpractice? Thank you. Yes, a very important question. AI is developing. faster than the detection tools can keep up with it. You're quite right to say that our reported cases of. detected and. proven AI. misuse remain relatively low. They're reported in our annual reports every autumn. But as somebody who visits.
schools and indeed colleges around the country on multiple.
schools and indeed colleges around the country on multiple.
11:56
11:56
11:56
11:56
Speaker 6
-
Speaker 6
Every time I speak at a conference, I'm approached by teachers afterwards. The line is always the same. We know because we know our students and we know their work because as human teachers, we interact with them. We know that they're using AI, but we can't prove it. We put it through the detection.
11:57
11:57
11:57
11:57
Speaker 6
-
Speaker 6
have a very strong and I think reasonably well. founded view that AI use by students is much greater. And indeed, when I undertake visits to schools and colleges, I always take the opportunity to talk with a group of students. pupils in the case of schools, and try to do that without their deputy headteacher being in the room with them and ask them to tell me. honestly how often they are using AI not to support them, just with their homework, just with their.
11:58
11:58
11:58
11:58
11:58
11:58
11:58
Speaker 1
-
Speaker 1
A three fold approach here that that we can take. I mean, we're doing a number of things already. I wrote to the chief executives of all the main exam boards. earlier this year to ask them to take this even more. seriously and to step up even. further their to checking for malpractice. But given what I've just said, I'm not naive about that. There will be limits within the current qualification system to what they can do internally.
We've just created a new post and stood up a new team on.
We've just created a new post and stood up a new team on.
11:58
11:59
11:59
11:59
11:59
11:59
11:59
Speaker 1
-
Speaker 1
Doing a dance, undertaking a PE . Performance. It's a drama performance. I think that's that's all much more secure. But extended writing coursework, we should avoid it. wherever we can. Where we can't avoid it, we should adapt it. So avoid, adapt, adapt it to make it as secure as it can possibly be. Bearing in mind what we're trying to achieve. And at the same time as all of that, we need to deter.
So we need to send very strong messaging to students about the. consequences and. sanctions of of.
So we need to send very strong messaging to students about the. consequences and. sanctions of of.
12:00
12:00
12:00
12:00
12:00
12:00
12:00
12:00
12:00
12:00
12:00
12:00
12:00
12:00
Speaker 1
-
Speaker 1
About the risks that some forms of. extended writing, non examined. assessment are now open to. There are always trade offs in assessment. You try to get one for a bit more of one thing. You have to do a bit less of something else, but that it is important that those making. decisions about reform qualifications are very clear what the risks are. And that's because they track right back to the fundamental purpose of .
12:01
12:01
12:01
12:01
12:01
Speaker 1
-
Speaker 1
There's a kind of. amplification of the views and. representations of those who own and design the technology. And that is certainly the case. In the case of AI, I think that the two examples I'd give are the use of AI by awarding. organisations for the setting of exam papers or the writing of exam.
12:02
12:02
12:02
12:02
12:02
12:02
Speaker 1
-
Speaker 1
I think, is an altogether more difficult question. Marking, by definition, has to happen behind closed doors. And if we were to. permit the. unaided use of AI for the. marking of. assessments and by AI use, I'm not talking about machine. marking where there is just a very limited set of possible right or wrong answers. I think that's okay.
And that's been happening for a while, quite securely without. concern by AI. In marking. I'm talking about AI. mimicking expert human. judgment over. sophisticated assessment.
And that's been happening for a while, quite securely without. concern by AI. In marking. I'm talking about AI. mimicking expert human. judgment over. sophisticated assessment.
12:03
12:03
12:03
12:03
Speaker 1
-
Speaker 1
Because I think there is a very real risk that bias could enter the. system if that were to be allowed. The fact is, we don't allow that. We've been very clear about that. AI can be used as an. ancillary in marking to support quality assurance and so on. That's okay. It can identify rogue markers whose marking is out of kilter with other markers. All of that is okay, but.
12:03
12:04
12:04
12:04
12:04
12:04
Speaker 1
-
Speaker 1
Evaluating the potential. impacts of these technologies on different groups? Yeah. I think the safeguards we've put in place are very good for making sure that we don't have differential impacts on different groups, because if we're essentially not allowing AI to do marking that, that has to be a primarily human activity. We're factoring AI out of those primary.
12:04
12:04
Speaker 1
-
Speaker 1
And then we have a whole other range of requirements that we place on exam boards to ensure that bias doesn't enter the market. The human marking process, for example, when markers mark exam questions, they don't know who the pupil is. They don't know their sex. They don't know where they're from. They don't know what school they went to. So it's possible to . Factor out
12:04
12:04
Helen Hayes
-
-
Helen Hayes
12:05
Speaker 1
-
Speaker 1
Policy about it, and faster than Ofqual's ability to regulate it? No, I don't I don't think that is the case. We are taking a very cautious and. precautionary approach with the the extent to which we're permitting AI to be used for high stakes qualification. So I don't think we are lagging behind what is happening. There may, of course, be awarding. organisations that operate. effectively in a competitive.
market who would like to do more, and that may be a reasonable aspiration or it may not. But my job, of course, chair, is not to do what the awarding organisations would like me to do. My job is to do what. Parliament's tasked me to do, which is to ensure that our. qualifications do give a reliable and accurate. representation of students.
market who would like to do more, and that may be a reasonable aspiration or it may not. But my job, of course, chair, is not to do what the awarding organisations would like me to do. My job is to do what. Parliament's tasked me to do, which is to ensure that our. qualifications do give a reliable and accurate. representation of students.
12:06
12:06
12:06
12:06
12:06
12:06
12:06
12:06
Speaker 1
-
Speaker 1
To really making sure that the integrity of. qualifications can be. safeguarded in the AI world. We are an outcomes based regulator. So our duty, our way of working, is to ensure that the results that students get when they take a regulated. assessment can be trusted and do accurately represent the quality of work that the student has. Evidence at the point where they took the.
12:07
12:07
12:07
12:07
12:07
12:07
Speaker 1
-
Speaker 1
is is what it is. So if awarding. organisations wants to use AI for data. analysis to help them. generate question papers or assessment plans, providing what we see meets the very demanding quality bar that we set, both when an awarding. organisation applies for. recognition and when it submits individual. qualifications for.
12:07
12:07
12:08
12:08
12:08
12:08
12:08
Helen Hayes
-
-
Helen Hayes
That simply doesn't work if you allow AI to do that instead. Thank you very much. And finally, as the committee considers this very big and challenging topic of AI and edtech in education, is there anything else that you would like to draw our attention to that we should be thinking about seeking evidence . On
12:08
Speaker 1
-
Speaker 1
and considering as we make our recommendations to the government. Yes. So I think this might be partially a recap of some of the points that I've made before, but if you'll permit, I'll just restate them. Now, I think we this is not a question of being for or against AI. It's about making conscious. decisions about where we as human beings in our education. system want AI to be.
deployed for the benefit of young people, for the benefit of education, for the benefit of.
deployed for the benefit of young people, for the benefit of education, for the benefit of.
12:09
12:09
12:09
12:09
12:09
12:09
12:09
12:09
12:09
12:09
12:10
12:10
12:10
12:10
12:10
12:10
Speaker 1
-
Speaker 1
Of human. learning to a machine. And I don't think we should do that. I think that that's a. fundamental moral question. And then just finally, to finish, chair, there's a question of accountability. You can call me for an accountability hearing at this panel. If my work was. outsourced to a bot, you wouldn't be able to do that. And I don't think that's right in a democratic society.
So this is .
So this is .
12:10
Helen Hayes
-
-
Helen Hayes
A high order, very significant question. Thank you very much indeed. That's been very interesting and a lot of food for thought for the committee. If there is anything else you'd like to draw to our attention, please do write to us as we progress with this inquiry. But we're very grateful to you for being with us, including short notice. I think it was as well. Today, that brings our evidence .
12:11