Loading summary
Transcript69 lines
- [00:02]
A
Welcome to the LSE Events podcast by the London School of Economics and Political Science. Get ready to hear from some of the most influential international figures in the social sciences. Welcome to everybody in the room and online. I'm Sonia Livingstone, I'm from the Department of Media and Communications at lse and I'm delighted to welcome you to this Evening's hybrid event, EdTech at the Crossroads of Pedagogy vs. Profit. Do please put your phones on silent because this is being recorded. It'll be a podcast later and I'm going to make some introductory remarks and then I'm going to introduce our speakers that you see here. So this event is being hosted by the Digital Futures for Children's center at the Department of Media and Communications and with the Five Rights Foundation. We support an evidence base for advocacy. We facilitate dialogue between academics and policy makers, and we seek to amplify children's voices. And tonight is indicative of our approach, we pose a problem. What's happening with EdTech in schools? We ask a child's right to educate, a child's rights question. We what is the right education anyway? We review the evidence and that's going to be our next report as well as what we talk about tonight. And we consult children and we've consulted around 450 children over the last few months about what they think of the ed tech in their schools and in their classrooms. And then we think about the policy implications. And that's part of what we want to discuss with you here today to tell you about the research and to raise some of the questions. So of course pedagogy versus profit is there for rhetorical effect. We do want to stimulate discussion. So pedagogy is variously defined as the art and practice of teaching. Children learn everywhere, but especially at school. At least that's our hope. And that's why countries invest in education. That's why education is seen as a driver of development and as a fundamental human right. But what happens at school? How teaching should be practiced and how children learn best? This has been long argued over and now the argument also includes the role of technology. Technologies of many kinds are being increasingly adopted in schools, increasingly AI enabled, now attracting considerable venture capital, millions, if not billions. And it's attracting a lot of critique. Is EdTech a tragedy? Asks UNESCO. Is EdTech useless? Ask the economists. Is it taking teachers jobs? Ask the trade unions. Is it wasting kids time or undermining their well being? Many parents are asking this these questions. At the Education Secretary's keynote last month, she claimed that Bridget Phillipson claimed that AI has the potential to transform education on a scale comparable to the printing press. She highlighted AI's potential to support every child's learning, particularly disadvantaged pupils and those with special educational needs. She also emphasized that the government is aware it must take its responsibilities seriously. It's committed to using AI in education in ways that are safe, effective, ethical and will support, not replace teachers. Just what policy that involves and what practice will result. This waits to unfold. So despite widespread use of edtech in schools, there are currently no standards or regulatory frameworks governing its procurement or use beyond compliance with data protection, which is a story in its own right. And responsibility with overseeing classroom technology is largely rested with individual schools, supported with guidance from the Department of Education. There have been recent calls to create a statutory code of practice to set minimum pedagogical standards for edtech in schools. The government is concerned this would be too rigid. The Department of Education has set up an EdTech evidence board which is developing a framework this is yet to report. Will report soon. The Information Commissioner's Office is planning a code of practice on the use of data in edtech. This is also to come. And just last month the government launched its non statutory safety standards on generative AI for edtech developers. So a lot is happening in this country, actually a lot is happening in many countries around the world, but we are focused today on the UK and it's a fertile moment for research. So our project, I'm just going to. This is part of a project called Better EdTech Futures for Children. The Digital Future for Children Centre always asks what good looks like, what could be better? And at this point we are aiming to stimulate a national conversation. What can work, what would be helpful? And it's been conducted jointly with the Five Rights foundation, which is a charity to support children's rights in the digital world. And it has funding also from the Garfield Weston foundation and the Rothschild Foundation. All our work, as I said, comes with a child rights framework. We work within the structure of the Convention on the Rights of the Child. That means we are very obviously concerned with the rights to education, which are articles 28 and 29, but everything that child rights is holistic. So we're also interested in the child's right to privacy, the child's right to inclusion, non discrimination, no commercial exploitation, expression, safety, well being and so forth. And in that guise, I just wanted to very briefly, before I introduce our speakers, note that we've been doing research in this field for quite a while and issued a series of reports which have Been really trying to explore.
- [07:01]
B
What.
- [07:01]
A
Are the emerging good practices and where are the problems. Especially originally focused on the question of data. What data is being collected as our children go to school and gradually hearing a series of problems in a way about the schools being the David to the Goliath of big tech and we worry whether they are losing rather than winning in that struggle. But we also have found ways in which schools are excited about EdTech finding creative and valuable ways with many hopes for it. We took a look at how other countries are handling some of these issues with we ask whether children's rights are being respected and the different products that they use. And we are now today especially going to focus on what children say and what is happening in classrooms. So I've used this phrase EdTech too many times already before offering you a definition. This is not. Well, this is a definition that will is doing for our working for our purposes. Right now we're trying to focus on schools. We're trying to focus at the moment in a whole range of different technologies because we're waiting to see which ones matter and they are changing all the time. The photos here are from bett, the annual trade fair which was just held in London last month at which there was an enormous amount of excitement and a very large number of companies there looking to advance both profit and we hope, also pedagogy. And it really raises many questions about what we think is happening. So I just want to acknowledge the DFC team before I continue because there are many people who have worked on this work and I want to introduce the speakers who are now going to tell us from their diverse kinds of expertise and really you can see them there. But we're going to begin with Dr. Sandra El Jamail who leads the project and is working, working with us full time, exploring children's rights, education and play across diverse digital and non digital contexts. And as you can see, she's also especially interested in hearing from children's voices and amplifying marginalized voices. Julian Sefton Green is, and Sandra is going to present our research and then the next three speakers will, will offer their response and their perspectives. So Julian Sefton Green is a professor of New Media Education at Deakin University and a lead researcher in Australia's Centre of Excellence for Digital Childhoods and has been researching in the field of education and digital for a long time and he's on our advisory board here. Rhys Spence, who will speak next, is head of Platform and Research at Bright Eye, a learning and work focused fund which is investing around upwards of 200 million in EdTech, I believe the biggest EdTech investor in Europe and was formerly director of the Education Policy Institute in London, as you see. And Jen Person is founder of Defend Digital ME and a member of the European Commission Special Group on Age Appropriate Design. She's worked also for the Council of Europe and was part of the inspiration when we began this work, thinking about where education data goes and does it indeed serve the interests of pedagogy or profit or perhaps both, if they are compatible. So we're going to hear from our speakers, we, we're going to make sure that we leave time for discussion. We're going to take questions then from you here in the room and also from those who are online and I believe and we will conclude sharply at 8 o' clock and I believe there will be soft or alcoholic drinks outside for those who would like to stay and talk more informally as we about the issues raised. So Sandra, I pass over to you.
- [11:43]
B
Thank you.
- [11:50]
C
Great. So, hi everyone. Thank you. Sonja. Thanks everyone for being here and thank you for everyone who's online joining us today. My name is Sandra and as Sonia said, I'm a researcher here, the Digital Futures for Children's center and I'm leading on the Better at Tech Futures for Children project. And I'm just going to introduce you quickly to our project and show you some of the early findings that we've been coming across. So quickly. What we did was between October and December 2025, we visited five primary schools and six seconds schools across the UK. So we went to Wales, Northern Ireland, Scotland and England, different schools around the country and different areas of the country and did workshops so child consultations with over 450 children. We also did 20 interviews with teachers and other education professionals. We tried to get diversity on recruitment. So one of the schools we went to was a school for children with special educational needs or children who are out of mainstream schooling for one reason or another, academies, faith based schools, Google reference schools and I think importantly schools that had different views on what how ed sex should be used with children. I didn't say the children went to the ages of 6 and 17. And we actually got to do consultations with children across the whole age range. So we designed two types of workshops, one for the younger children and one for the older ones for the younger children. I think for both groups we really wanted to understand what children are using. So what products, what devices, what apps, how do they use them, what do they think about them, do they help their learning? Do they hinder their learning. And what would they like to see done better? What change would they like to see? So for younger children in primary school, we designed a few different methods to collect data with them. We did drawing as a participatory method, we did vignettes and we did something called the magic bag approach, which we drew on from Lerna Arnold's work. And so basically, we cut out different logos of apps that children would be using in school, different devices, and went around the room as children to pick one logo out of the bag or one app or one device, and ask them, how do you use this? What do you do with it? What do you think about it? And we gave them these stickers which worked really well. I'd recommend it, happy and sad faces. And they could show us what they liked, which apps they liked, which they didn't. And we had a great discussion around why. So for older children, we did something slightly different. We asked older children to write down what devices they used in the classroom, what apps they used for classwork and homework, Gave them the same stickers, which they also left. And there was lots of great discussion and around what was good and what was bad and what they liked and what they didn't. We asked children to draw screenshots of the apps that got the most or the least happy or sad faces and to write down what were the features that they liked or didn't like. And again, lots of discussion and we asked children to write down, what are things you think we should keep, what do you think should be changed and why? So some findings. There was a lot of variation in how edtech was procured by schools. I think a lot of it came from management down. So management made decisions and then it trickled downwards towards the teachers in the classrooms. Some schools accessed apps through, like a government, like in Wales, they have this help website where you can access all the different apps that are used in the classroom. Some schools have access to like London Grid for Learning, but also provided them with lots of apps and they often considered those to be safe to use. Didn't really question much what they were using there. Teachers, I think, wanted to have a say in what was being used. They did lots of research. They often thought apps to use with children through word of mouth, through Facebook groups, LinkedIn, Twitterverse. They also went to Bet that Sonia mentioned earlier. So this big conference or this big showroom where there are all these different edtech companies showcasing their stuff, so they'd get ideas from there as well as from subject association websites and they found those to be good sources of information somewhere. Google reference schools or like use other big tech products. And I interviewed the teacher at the Google Reference school. She was saying how they got lots of training from Google. They would fly them to Madrid, they'd fly them to the us. The children could even go to the US and be involved in all this training on what are the latest updates on Google products. And they also considered information to be secure and safe. Lots of variety in devices, apps and policies across the schools. So some schools had drones, NVR headsets and robots, 3D printers, others didn't. I think across the schools, generally there were Chromebooks or laptops and PCs and interactive white. Again, there was variety in whether the child like the ratio. So was it a one to one ratio where the child had their own device that they could take home, do their homework on, or was it shared where the child then had to do their homework on a smartphone on a shared computer or do their homework in school in the morning in the PC labs? Policies varied a lot. So some schools consider, consider AI unsafe for children. They only use it rarely and they could only use it under adult supervision and like ICT class. Whereas other schools considered it was their responsibility to guide children on how to use AI safely. They were already using it and it was going to motivate them. So there were positives that they found in that. And apps varied greatly and often budget was a big determining factor of what was used. So like in one school, they really didn't have the finances to pay for many apps, so they often encouraged children to use free apps or free versions of apps. So they only got limited features that they could use. This limited what they could learn and what they could do. Children sometimes signed up for a free trial so they could get the full range of features, but that meant that they risked forgetting to unsubscribe or, you know, cancel the free trial and. Yeah, what happens there? Conversely, I think another sort of example of, you know, apps that were used, there was a school where they got funding from the local authority, but they also applied for grants for tech. They did research with universities where they got funding for tech. They contact companies and ask them if they could get free licenses and they were often given those licenses and they found that they wanted their children to have the industry standard because that's what's best for the children. So how does all this variety influence different children's learning opportunities? Yeah, and they're reds. I'll skip that. AI was in a lot of things, including homework apps. So I think teachers used AI for support with lesson plans. They used it for setting homework. They use it for automatic, like grading of assignments and homework. They found that it gave them a lot of data on children which they liked and they found very helpful and they felt that they can then improve their lessons and tailor them to children's needs based on this data. I think they were, and they came up in interview that they were also aware that the apps they like aren't necessarily the apps children like. So, for example, children reported many issues with AI with homework apps. Sorry. So they reported lots of glitches, lots of software freezing. They sometimes lose their work that they've been working hours on because the, I don't know, the Internet goes or there's some glitch in the system. They sometimes couldn't log into homework platform so they couldn't actually do their homework. And when it came to any AI and homework apps, children had a lot to say. They found that AI generated homework or assignments were often unclear, so they couldn't actually understand what the question was. They couldn't understand what they were meant to be doing. They found that tests were focused too much on giving an exact answer to something. So for example, if I write the correct answer, but I have the wrong punctuation, I have the wrong spelling, it's also automatically wrong. So it didn't really capture their thought process how they came to the answer, which I think is an important part of doing that of homework or classwork. They also said there was a lack of feedback and this upset them. I think it really frustrated them. So they get an answer wrong in homework. They might get a video saying, oh no, this is how you should do it. And they don't really understand the video. And there was just a lot of frustration. Some children mentioned crying or knowing of children crying because of the frustration of just not understanding what they're meant to be doing. And children often turn to AI, they often turn to like things like ChatGPT, Goth AI, Snapchat, AI to try to find guidance when the AI within the homework apps wasn't giving them that guidance. So yeah, they tried to get step by step guidance or trying to understand step by step how to come to an answer, why they made a mistake and how they can fix it. And they said that they sometimes went to AI when they felt they didn't want to burden a parent or a teacher. Timers were embedded in apps as well. So when it came to homework children, sometimes there was a timer that was tracking how long it took them to complete their homework. Children were really frustrated by the timers because they felt they weren't accurate. It was unreasonable to have them. They didn't count the time properly. Sometimes you have to sit there and keep moving your mouse to make sure the timer doesn't stop. So it was just like they were trying to navigate the system and try to find a way around it. And I think importantly, timers couldn't be adjusted for children's individual needs. So it created a lot of stress, a lot of anxiety, especially for children with special needs and special educational needs. And this goes against the claim that edtech is personalized because this is a case where it's obviously not. The last hometown think I'll make is on gamification. So a lot of edtech was gamified. Most children enjoyed it. I would say they thought it was fun. They liked the competition, competitive element. They usually like leaderboards. But what about children who are on the lower end of the leaderboard who are always getting the lowest scores, always being compared to others, not only in the same school, in the same class, across the country and other schools, maybe even across the world. Also, there are questions around motivation. So is children's motivation to learn or is it to win? And you can see there's a comment, it's bad if I'm not first. So last slide. What would you change? We asked children what they would change. They had a lot of to say and these are only a few things that they want to change. Some children mentioned wanting reminders before free trials came to an end. So they didn't have to end up paying a subscription fee. They didn't want advertisements in apps, they wanted homework timers removed. They wanted less closed ended and like really simplistic AI generated questions for homework. They wanted clear explanations why was something wrong, and they wanted clear guidance on how to get to the right answer. And I think they wanted close, interesting relationships with their teachers. They wanted teachers to rely less on AI generated slides and PowerPoints. They wanted more collaboration, more participation, and they wanted to be able to contact teachers directly for help when they needed it. So that's a bit about children and teacher said. I think we'll go to the panel now for some questions and continue the conversation. Thank you.
- [25:30]
A
Santa. Thank you so much. And I'm going to pass now to Julian who is going to talk about his research and his response.
- [25:41]
D
Thanks very much, Sandra. Thank you very much, Sonia, for inviting me. So in the past I've researched the whole use of digital technology for learning and leisure, which has also Involved looking at portals and platforms that support homework, just as Sandra talked about a bit. But especially focusing on the growth of homeschool communication platforms and also the ways that digital technology now supports what we various kinds of learning outside the school. I'm currently based in the university in Australia, as you can tell from my accent and a couple of bits of projects that are going on there. First of all, researching the scope and range of the edtech industry for under eights in Australia, especially focusing on the platforms, the apps that families purchase for what they think is their children's learning. And secondly, we're developing an interest in how edtech is made, so its production culture. So focusing on the companies, the people, the labor practices, the kind of careers of people in EdTech as well as the intermediaries and networks, because there's a lot of working around edtech to get it into schools, which isn't within the companies, but exists in a whole sort of hinterland of businesses. So they're very interesting things. But edtech as a term or phrase describes a range of technologies, processes, practices and developments. It's not really a very discrete field in some respects. So a learning management system, or lms, is very different from Teach yourself to read app. One of my favorite studies of edtech is by a colleague called Antero Garcia, based in the us, who studied the school bus as an educational technology. You know, in America, the yellow bus is because it enables access to education, it sorts, it stratifies, it manages children's time. So the school bus is a kind of ed tech, even though that's probably not what we're going to talk about today, but just to sort of give an indication of how we need to think of the concept in a very elastic sense. There's always been technology in education, right from the Plato steps that you came down to get to this hall, which are apparently modeled on a picture of Plato's Symposium or something. There are always ways in organizing classrooms, in organising slates, chalkboards, you know, we want to think about. You have to think of Ed Turk within this history. I think at the moment we're interested in two key change processes, two really important social transformations, and that's what's energized the debate today. The first of these, and forgive me for using ridiculous academic phrases, I'll try and explain them decently. The first is what I've called the platformization of education. And the second, and this is worse, really, is the pedagogisation of everyday life. So I'm going to explain what I mean by those processes, because I think they underpin and they drive what we're talking about. The platformization of education refers to the increasing role of privately owned digital platforms that determine, coordinate and define what happens in schools. Teaching and learning does not just take place between people as a relational or intergenerational exchange, but now almost always through or in some ways connected to a digital platform. And the basis of this platformization is simply the power to reduce all human activity to a measurement, whether we're talking about attendance, learning, progression, well being, mental health assessment, school lunches, whatever. Right? In other words, all kinds of school management, learning, peer activity, teacher workloads, school lunches, etc. Now take place through a series of either, as Sandra's just brilliantly demonstrated, badly coordinated, half broken, or perfectly brilliantly working interoperable platforms, changing what it means to be at school or to be a teacher, or to be a student, or to be a parent. And this raises a series of questions about whether teachers like, you know why people go into teaching, is it to become a platform administrator? I don't know. It raises questions about privacy and data. It raises questions about whether it's good for learning and whether it's even good for managing school. Of course, there's a history here of researching edtech, and the key themes probably are around the unease we feel about the role of private companies within what's traditionally been a project of the nation state. Education is fundamentally central to the modern nation. We're also uneasy about the metricization of everything, of turning everything into a number, of treating every kind of human transaction in numerical terms, and of course, the challenges for trust and privacy, data and rights, and particularly where it begins to blur what used to be pretty firm boundaries but now between the home and the child, or the child within the family, or even the child as it now exists within the school where the school is expected to act in loco parentis. The second theme I mentioned is what I've called the pedagogicization of everyday life. Forgive me, the term has a number of meanings, but it relates to the ways that schooled definitions of learning. So knowing your times table, repeating things by rote, whatever, it doesn't really matter that schooled definitions of learning influence the whole range of all kinds of learning in the home, as well as theories about how we might be changing in terms of our identity, and particularly in terms of discipline, values and achievement. And I think this is central to the crossroads which lies at the heart of the panel ambitions today. So behind this, I think, is increasing competition for well paid jobs, there's a broad decrease in social mobility, the changing role of education as a route to social betterment, social mobility. And so that as a consequence of these kind of external pressures, there's an increasing intensity on child rearing, an increasing pressure in the home to emphasize forms of educational attainment. I think the argument here is that the growth of ed tech, especially for young children, and the attention to supporting and developing learning is best understood in terms of the exploitation of this increasing anxiety. Either exploitation might be the wrong word and it might be the generation of this kind of anxiety. We just don't know about these larger contexts because families are increasingly turning to private resources to help them and their children achieve more. Again, this process of increasing penetration of school types of learning into the home draws attention to measurement and the ideas and that all achievement and all education can only be defined through the forms of entity attainment offered by EdTech in their home. I'll stop, I'll stop there.
- [32:46]
A
Thank you.
- [32:51]
D
Okay.
- [32:51]
A
Lots of challenges, lots of big words. Rhys, so you speak from a different perspective and we would love to hear your thoughts.
- [33:02]
E
Okay, yes, thank you. Thank you for having us. Firstly, really enjoyed the line in Sandra's speech around children's dislike of AI slides, which I thought was really interesting because not only can they discern that it's AI, but they're also thinking critically enough to realize that it's probably not what they want their teachers to be focused on. So really appreciated that and I think it sort of highlights something that I'll go on to talk talk about a little bit. So at brighter, We've invested in 50 companies so far and we're just in the process of trying to invest another 25 over the next two years. I would say around 10% of those companies, around 5 are actually focused on K12 in the school age realm. And that includes companies that are selling directly to parents as well as selling into schools. So to highlight two companies just so you get some more context on us, one is called Stewie, which is the largest school operating system in Europe, operating in Germany, Austria, Spain, a number of other markets, is doing really, really well. They basically have a suite of tools that help teachers save a ton of time, help administrators have a better grasp of the data that's running this that they can use to run their school and also creates better relationships with between smoother relationships between parents and school leaders. The second company that I'd mention is Zeneducate, which is a supply teaching marketplace. On one side they have supply teachers. On the other side, they have schools. And because they run more efficiently than traditional recruitment agencies, they save schools money and they pay teachers more. So, you know, a really, you know, elegant solution to a problem that is, you know, a big deal here, a big deal in Europe and a huge deal in the US which is why they're working there quite closely now. It's generally quite hard for us to back companies in the schools market, which I think is a good thing and a bad thing. It's a good thing in the sense that I think it reflects that schools and teachers individually think very critically about the solutions that they choose to bring in and therefore it makes it hard to sell into those schools. Sales cycles are very long. Finding the right decision makers within schools is really, really hard for companies. And I think ultimately, I think interoperability of different solutions and how they work together is a real challenge as well. And so I think, I mean this isn't the subject of today, but like if you look at the access to AI tools that teachers have now that every, you know, all of us have, ultimately what I think that that will remain result in is more localized solutions, more specific solutions that schools develop for themselves and for their local communities based on their context. And I think, you know, you look at, I don't know how familiar you are with platforms like Lovable and the fact you can build anything you want within, you know, you can build an mvp, a minimum viable version of something within a couple of minutes. But you know, you think about what a teacher will eventually be able to whip up, you know, almost instantly. And so that makes it really hard for us to back them. And that's because our model is that we need companies to be able to get to a scale of 100 million in revenue per year plus in order for our model to work. Like how many, how many companies are realistic selling into schools are realistically going to get to that level? And the answer is very few, which is why we don't back many. And so that's a, you know, that's a interesting thing. I guess another, another trend that we see within founders is that they start off very focused on sort of the casefile mission. You know, they're often very mission driven individuals that start these companies and you know, very noble aims and then they'll come to investors like us and then some of that sort of initial intent might be lost and then they might end up realizing that in order to sustain themselves and build a viable business, they have to sell into corporate environments instead. And we see that a Lot you see a good example of that would be Sana, which is the Swedish company. Their founder started the company, I think, when he was 16, 17. Incredibly impressive, incredibly smart guy. He's trying to change the Swedish school system and then what does he realize? Realizes that they're not ready to buy what he's selling, and that's fine. But he took investment and what that resulted in was him pivoting his solutions to selling into corporate environments. And, you know, he's just sold his company for 1.1 billion, which is insane. The one positive thing I'll say about this, and I don't want to ramble too much about Sana, but they're cool because they've done so well. They now work with the Swedish government to distribute their AI tools within schools, within Sweden, and also to government departments as well, that would potentially find, you know, they're upskilling all of us effectively. And so it's nice that they get to return to that mission. And I guess, you know, it's a rare case, but it's an interesting one to talk a little bit briefly about, you know, the impact side of all this stuff for us. You know, I think the biggest issue that schools have when they're choosing between solutions is the companies that market best. Their solutions will often win. They'll win the contracts. The companies that have the biggest, existing, strongest brands will also win the contracts. So you look at, you know, if I'm a teacher and I'm short on time and Pearson comes up to me with a new solution in X area, and then there's also 10 other startups that claim to have better impact, but I don't have time to verify it. And there's no standards that those solutions are held to. Which one do you think I'm going to go for that? You go for the Pearson. And so I think it's a shame, really. And it does sort of stifle a lot of the innovation that we see in the. In the schools environment. And again, that's another reason why I think we have struggled to find really great opportunities in the space. I guess just to give you a sense of what we look for in impact perspectives, there's four ways that we look at the impact of companies that selling into any environment, including schools. Our rubric is we bet companies that make learning more affordable, effective, efficient or relevant. Affordable, obviously saves, in this context, saves schools money. Effective, improves learning outcomes. Efficient saves teacher time and more relevant, better connects students to onwards sort of career and training opportunities. Opportunities because we back early Stage companies, the companies we back, they're only 10 people or so. They're very, very small organizations when we first invest, and then they grow. Ideally, that's what that means in practice is that they, they probably don't have all that much time to have conducted an rct. You know, like they, they're thinking about the next six months, not, not the next sort of three years. And so we try and keep, you know, we try and keep them to task on an impact approach where they report to us against the SDGs and some other sort of impact goals that we are internationally recognized. But we're also kind of realistic about what level of proof they're likely to have when they get to us. We do support them very closely with getting in touch with great impact mentors and people that are able to support them within the independent study when they're ready. The final thing I'll say, and I'm aware I'm not really stuck to script, is basically I was asked to mention the fact that in the US there is a system of validating the quality of education interventions that sell into schools, which is called the ESSA standards. The ESSA standards. There are four levels to the ESSA standards of varying evidential evidence and rigor of proof that the product does what it says within their requests for proposals. Within schools, when they're asking for new solutions, they often have a set level of ESSA standards that they basically will ask for, which is really interesting. And like, I think if you ask lots of academics in this country about the ESSA standards that are familiar with it, they hate the ESSA standards, but I've not really, really understood why. And also if you ask the Americans, they don't understand why either. So take that as you will. But yeah, that was a ramble, but hopefully useful.
- [41:26]
A
No, no, super useful, thank you. And yes, I wanted you to mention it because we've been kind of calling for standards here to deal with the kind of incredible variation that we're hearing across schools and also the uncertainty that we're hearing on all sides about whether the right quote solutions are being invested in different schools. So interesting to know that there are standards, but when you get them, people hate them. So it's a tough one. Jen, over to you.
- [41:57]
B
So the thing that struck me most from your research, it's fantastic that there's proper children's voice in this now. I think we would probably, if we were at the BET show, have had a completely different perspective from the children that go to that having been coached for weeks Beforehand, which I was really surprised by. But they get coached to go and be sort of in the audience and take part in ed tech on the stands and take part and show on and do their demos. And you see this sort of enthusiasm and fervor for the shiny, exciting looking toys. I've been going now probably for almost 10 years, almost every year. And it's what's fascinating to me is that actually the show has got in my view slightly smaller. There are fewer robots than there used to be. There is less and less, I think focus on sort of generic, sort of this will great big goals for what we'll bring about in teaching and learning. And there's still a very strong core of administrative tools. So it's really focused on teacher school admin still. And I think what your research is pointing to and as we're finding across Europe, feedback from lots of countries is that there is less and less evidence that actually this brings about any sort of educational outcomes genuinely.
- [43:24]
A
So.
- [43:24]
B
So I'll start with the child, I think, as you did, and I was asked to speak a little bit to why perhaps where we've got to where we are now, especially around the sort of data protection and legislative perspective, why laws aren't possibly working and what we can do about that. And I think it was fascinating to think when you look at this from children's perspective, you're listening to their voice and what are they saying? I think why we have got to where we have is there's also a time lag and we need to think about this also in the sort of political big picture context of what has been going on in schools over the last decade or 15 years. And the most striking recording I found on YouTube was 2013 when Michael Gove gave a talk. He was then Education Secretary. We've been through a few since to the National Summit for Education Reform and it was an organization established by Florida Governor Jeb Bush. And in that speech he explicitly links education reform and edtech to the knowledge economy. And he's very keen. As you were talking about, this is productivity. You've got to be effective, you're going to be productive, you're going to achieve something. Go back to Sandra's research and you find the children are being tasked like Amazon workers. If you walk away from your desk, you get punished. The time time's out and how does that make them feel? And I strongly suspect that we have a decade of children. Everyone's desperate to point to social media being this great harm, but I strongly suspect there is a really strong correlation, if not causation, with how we are making children feel in the classroom using these technologies. I've had three children through this education system in the last 15 years and what has changed? So I'll give you three case studies, some really shocking in 2010 and you can still find the story on the BBC. If you look, there was an edtech platform called Sparklebox. It's now been rebranded, it's got new owners, nothing to do with the previous people that ran it. But that person was a convicted pedophile. He was charged twice with offences and he was running this school platform. What has changed in those 16 years? Still, as far as I know, there are no checks on who can run an edtech platform.
- [45:50]
A
Hi, I'm interrupting this event to tell you about another awesome LSE podcast that we think you'd enjoy. Lseiq asks social scientists and other experts to answer one intelligent question like why do people believe in conspiracy theories? Or can we afford the super rich come check us out. Just search for lseiq wherever you get your podcasts. Now back to the event.
- [46:18]
B
What about the technology that's come in in those 10 years? Well, it's become more and more intrusive. You're asking children about more and more personal incentive things and how it makes them feel is also being analyzed by some of these ed tech. So you've got tools that are being used to ostensibly claim they measure well being and mental health and they're marketed to school, to schools as, you know, identify how many of your children are feeling whatever they're feeling and what you can do about it and then might sell you upsell your program afterwards to treat their well being or their mental health. But let's put in the small text that this is not a diagnostic tool. And why? Because if they did, they'd be regulated by the mhra. So there is no health regulator involved in these apps which are ostensibly being used to measure, regulate, profile and potentially treat or intervene in children's mental health and well being in schools that I find really shocking. So why have we done nothing in the last 10 years? And I think one of the reasons is that we are hands off in education. And Sandra, you found, you know, these different devolved nations are doing things differently. Scotland has a different approach because they still don't have academies, so they have local authorities that still do data protection assessment, for example. And as part of that they do that regulatory check before they procure something. And they know they have a sort of, you know, checklist of what's green and what's red, what products are approved and what are not. Wales has the Hub, so they have this quality check. Interestingly, I think in lockdown prior to Covid, parents could object to the use of Google because it stored data abroad at that time it was an optional extra. Some of Google's pieces are not core product, which means they're additional product, which means they do additional processing of data which goes to Google's choices how to use it, which in my view makes them a controller at very best a joint controller, not just a data process processor. As we keep talking about education, after Covid came in, the option was taken away. You had to use Google if your school was using Google. And so this speaks to kind of props to the piece that then is who else is left? Who else is affected by this? Parents. So parents I think have the right under the UDHR in Article 26, 3B to also be involved in the choice of education that their children have. Now where are we going politically with this right now? Is that everything everybody is speaking about how children are feeling and how parents are making finding their children feel. Because even some tools, for example, like parent pay, if you don't have the technology or the apps at home to be able to use that tool, which your entire school cashless payment system functions through, you can't take cash to the school if you don't have a phone. Or you may choose not to use a phone or a laptop, right? It may not be a digital barrier, it may not be a choice because of economics. It may be a choice you have to take a printout of what you have to pay for in school to somewhere that does it like a co op or a supermarket or a shop. So parents are being sort of given this back office tasks that used to be within a school as well to manage. But they're just choices. Their pushback is coming. So to just touch very briefly on a couple of things in data protection law that we should be saying are working and are not, and I'll wrap up with that, is where are the laws that we have today and why are they not working? Data protection law, even if you're in a non consensual environment, so an environment in schools, children may not choose to go there, they may choose to be in education, but obviously they as children, children perhaps don't make a freely given choice. Schools are using data protection law but completely ignoring all the rights that are associated with it. And what is missing from some of that is the ability to exercise it. But data Protection law is not enough. We need to be looking at public sector equality duty. And I see Tracey here in the audience tonight from data tech and black communities who did outstanding work in Birmingham and other places that really needs red. If you haven't seen it, go and look at it. They did terrific work finding out about public sector equality, duty and dignity and the human rights sort of frameworks we had in schools there. And we need to be joining up these different pieces of legislation to say what do we have in law? Why have we just completely ignored it in schools that we've got to the situation where we are now with this and then understand the politics behind it, the economics behind it and see how we put these things together for what we might achieve and why we bring about that change and say it is urgent. Children can't wait for policymakers to get their act together. We've been waiting over 15 years. My children are now out of the state education system and we need to get these things right for the generation that's coming who are seeing more and more intrusive technologies, what that's going to mean for things like identity, your use of your information that's coming out of schools, going into the home office and policing and other places around digital public sector systems and what that means for them when these big corporate companies are determining the values that goes into those systems and into those children. And like our first example of sparklebox means that actual strangers are interacting with our children in the classroom.
- [51:36]
A
Thank you. Heard some very different perspectives, some very informed and expert and strongly felt views. I'm going to invite questions in just one minute. I just want to ask the panel. I'm just picking up a little on Jen's question of how do children feel in the classroom? And this was something we've very much had in mind in the research. I'm conscious that in the Good Childhood survey of British children last year, of all the things that really put pressure on children and their mental health, actually social media is a bit down. Stuff happening with their parents is a bit low down. The number one thing that children are finding hard is academic pressure. And I think that sense of what it's like to be in our school systems and being implications for their learning. But before we come to, I just want to ask if to raise how we're all feeling, have we seen some good uses of edtech in practice? Have we got some example or are we super worried? And I just wonder if anyone wants to say what does good look like? Whether you've seen it here or perhaps elsewhere. And Sandra, I'm thinking we did see lots of kids to be fair, who were enthused that one day they might get VR headsets in the classroom. And we met some lovely robots.
- [53:11]
C
Yes, there was excitement over it. I think there was good use of edtech definitely it wasn't all bad and I think a lot of it had to do with how the teacher took on the role of using the edtech with the children. So there was a, a lot of creative work that that was done using and even the teachers were telling me they're like, they'd use things like I don't know, garage bands and they'd use Canva to create posters. They were making music and they were in like working together, doing group work and they'd come up with really fantastic like the teachers were saying even really professional looking stuff that children were very proud of. So I think that's a very positive.
- [53:52]
A
Good. Thank you.
- [53:53]
C
Yeah, there are others but that's one.
- [53:55]
A
No, no, that's good. Yes.
- [54:00]
E
Happy to go.
- [54:01]
A
Have a go on.
- [54:02]
E
So I mentioned the two at the top that I think have done really good work. I think the thing I would remember from our perspective is that if it doesn't work well and do what people think it should then it probably won't scale within school. And so that's why I think it's almost like yeah, I, I just, I don't think it will because people, people rightly will stand on their principles regarding how things are procured, how it makes children feel, etc. And I think the thing that we need to kind of, I think I keep thinking about at the moment is obviously we're seeing industries totally transformed by AI but like we just have to try and remember that we all have agency to choose what we use and obviously there are things that are put in front of our noses that we are supposed to sort of, you know, use in our day to day. But ultimately we have choose like we have have the choice and I think so to link those two points together. Like I think the solutions that will end up doing really well are the ones that are like sympathetic to the human experience and what we feel like we should be prioritizing. For example, it's engaging teachers in the way that they want to be engaged. So engaging parents in the way they want to be engaged. I think that there might be some sort of short term, this is very optimistic. There might be some very sort of short term negative impacts when like the wrong solution gets into the wrong hands and it's not good. But then I think in the longer term, I think it will sort of course correct. And that's partially because I think there will be more localized solutions developed that are specific to groups of people that need certain solutions or to specific geographic or local contexts, etc.
- [55:40]
A
I think children often don't get that choice. I think that is one reason that we wanted to kind of focus on what's happening in the classroom. They are told what to use and they better get on with it. I might actually turn to the audience, if you will forgive me, because I can see hands beginning to go up. I'm going to ask people please to say the name as they raise a question and I'm just going to take a few questions together and then those who are online, if you can put your question into the Q and A box and also introduce yourself at the same time and we will come to you. I'm going to start here and then here. Yeah, thank you. Don't feel you have to answer every question. Pick the questions you want to.
- [56:28]
F
Yeah.
- [56:28]
G
Jennifer Powers from the Unplugged Coalition. Risa, I thought what you said was really interesting and I get what you say about where Bright Eyes invests, but I wondered if you could just quickly say what you're seeing in the rest of the market. You know, who is investing in the classroom at tech, if anyone, what sort of things are they investing in? What you know, why have you decided not. Not to go there? I know you said money and scale, but like, is it because you don't think you that the efficacy can be proven or is it just the financial aspect? So I'd really love to hear your view on, on not the admin side, but the actual learning side of education. And then I don't know if this is a question for you, Sonia, but just how confident are you that we're going to get to a place in the uk or how long will we have have to wait before we can have whether it's not the ESSA standards, but we, we can actually have guidance to schools who are in this wild west of being sold stuff that doesn't necessarily improve children's lives and in many cases I'm sure makes their learning and their well being worse. And you know, every day this is happening to kids and I guess I'm impatient for change and I'm just wondering of your assessment.
- [57:44]
A
Thank you, thank you, thank you. We're going to, I've made notes and we're going to come here and then also at the front in the white yeah.
- [57:54]
H
Should I say my question or we're just going to wait for the answer?
- [57:57]
A
No, please say it. We're going to take them in groups, so if you say your name and your question. Thank you.
- [58:01]
H
Hi. So this question's for all four of you. My name is Francis Jones, an ex teacher who got burnt out for too much marking, ended up making an AI called snapgrade that does the marking for us.
- [58:10]
A
And.
- [58:11]
H
And we've been invited to tender from a bunch of mats and private school groups. But the issue about it is they want larger edtech startups. I've been having meetings with a few of our competitors for potential partnerships because we have a high chance of being approved for tender if we unite up together. But one of our competitors I was having a meeting with today, me and the founder had opposing philosophical beliefs on AI in education. My stance was AI should be used to augment teachers and support and I promote full human input and verification and transparency. But he believed that AI should be there to replace teachers because AI is more accurate and more reliable than human teachers and doesn't believe that human input into AI and education will have any preferable benefits. So I'm more into the whole AI should be transparent for teachers. He's more in the stance of AI should be a black box that we shouldn't have any input into. From your perspective, from an academic but also investment point of view, which one of us do you feel is more correct?
- [59:23]
A
Great, great question. Thank you. One more and then the panel will have collected their thoughts. Yeah, in the white here.
- [59:32]
F
Thank you. Hello, thank you for your speech. And I'm yaying from secondary school teacher. I'm still a teacher. I'm a language teacher and a math teacher. So I have seen like this is a revolution, AI errors is coming and so have a lot of impact, good impact in our students. Like sparse masks. We use sparse masks a lot in the school. So I have two one question for reason you mentioned about edtech for nowadays four elements like one is affordable, efficiency and the relevant. And I was wondering like relevant it's not only relevant related to the curriculum exam for Pearson or the AQA or whatever this, but how much the teacher's role need to be involved involved in this ad tech development or design because when you design, I think this edtech is not only about the business of the tech people, they are more about the educator or educational. Like professor, we know about the students and what the misconception, what kinds of difficulty they met in the classroom. So how much the the teacher should be Involved in the ad like this ad tech being designed.
- [60:52]
A
Yeah, great, great, great question. Thank you. Rhys, why don't you come in first and tell us how is the rest of the market you are addressing.
- [61:05]
E
And okay, so we, so there is a. There's kind of two, two bits I'll talk about. One is, is there are particularly in the US and we tend to follow the US on a lot of this stuff. Not in everything obviously but around supporting earlier diagnosis of like additional learning needs via not, not through necessarily sort of, you know, AI assisted sort of diagnoses, but more about technology that can accelerate parts pathways and create better information flows between what teachers are seeing in the classroom, what parents are seeing at home, bringing those observations together and then accelerating a pathway to meeting a specialist that can make a diagnosis of a learning need that then can be addressed within the classroom. And it's in that way, it's end to end in the sense that it leads from the point of sort of suspected learning, needle learning difference to actually having actionable support that can be used in the classroom to that end. There's a similar wave of companies that are offering for good and bad well being and mental health advice to students that are demonstrating signs of mental health. That's largely again in the US there's a number of companies. I think the leading one's called Cartwheel if you want to check them out. I'm not sure how I feel about those solutions because I think it might lead possibly to sort of an over diagnosis of a lot of these things. And also, you know, it's a very personal view but kind of introduce like therapizing of kids like earlier than is necessary when they're going through so many changes themselves emotionally.
- [62:50]
A
I think in a way the question is also is Bright eye following your personal view or is there a kind of reason why you invest here and not in that how as a company, how do you.
- [63:02]
E
We're a team of eight and so okay personal view is very close to okay our investment decisions for the right. I mean we see a lot of companies that are doing that kind of stuff and they're all, you know, very well intentioned stuff. Again, it comes down to scalability for us. You know, they are perfectly good and well meaning companies. Would we back them? Like probably not. Is the is the answer. The second category is rather than selling to schools, we see a lot of companies that are trying to use the B2B2C model. Basically means you're going, you're distributing two schools that are then distributing your solution to parents. And then parents can basically, you know, buy into as much of the solution as they want based on their child's preference. And you see a lot of things in there around, you know, additional math support is a particularly positive one. If you go even earlier than sort of school age. There's a lot of apps that are focused on building curiosity in children at the moment and storytelling skills, problem solving skills, lots of things that are non subject specific. So they're the two sort of things I'd highlight.
- [64:02]
A
Thank you, Sandra. I think a lot of what we saw really did make the case for AI to augment, not replacement the teacher. But I'm just wondering what you're.
- [64:13]
C
I need to give examples because I think it came up so much. Children are really frustrated when they couldn't understand what they were being told. They couldn't understand why they were being given a negative mark, which is usually like there was one app where you go on to see your homework, to see your timetable and to see your behavior, behavior charts. And the first thing you see every day when you open up this app to see your timetable, the first thing you see is your behavior and you get all these negative points. And I think there must be like a drop down list or something or just AI generated and it just gives a generic answer as to why you got these negative points. And children just didn't understand why, why they were being given these things. They couldn't understand the story. Sparks Maths, what do you call it, like feedback that they were getting. They couldn't understand why something was obviously AI not. It wasn't working well. And I think children should really be consulted on this because they're the ones using it. It's affecting them.
- [65:20]
A
Or Sparks Maths was that was. You got it wrong. Watch a video.
- [65:24]
C
Watch a video. There was no real.
- [65:26]
A
Doesn't explain.
- [65:27]
C
It wasn't necessarily a video on that specific question. It was kind of a generic video. And it just didn't explain there was a thing with AI slides. It's not exactly what you're asking, but I just. Reese, you mentioned something at the start about the AI slides. And I think children, it's not that they just could tell immediately that it was AI generated. They could tell that the teacher didn't understand the content and so they couldn't understand the content. And it really upset them. Like they wanted to learn and they wanted teachers who were engaged and were really passionate about the subject and wanted children to learn. And I think this was usually for cover teachers. It was more of a thing Like a cover teacher just creates something quickly and so AI kind of gives you this shiny looking thing without really the substance, at least in those cases.
- [66:20]
A
Thank you, Jen. Are we still in the Wild West? Are we going to wait forever for better standards and regulation? And do you see it actually coming in the next month or year?
- [66:36]
B
So what are we waiting for? I think, right, so some of the things I think keeps coming back is we seem to be searching still for a reason, reason to put these tools in front of children sometimes. Like wait, we still, we have all these things that come up and you know there's a brilliant, I would say go back to good examples. There's a brilliant teacher, real life teacher, who uses recorded videos to explain maths, but he introduces every single one with his strap line. Do you know how annoying that is? When your child is sitting in the living room doing their revision for a GCSE exam and every 5, 10 minutes they have the same teacher introduce themselves with a 30 second intro. They're not designed for the child, they're not designed for the purpose for which they're put. So we're not replacing a teacher there, but we have to say what's the purpose of a teacher? And we are moving in some of these AI particularly driven models to seeing teachers teaching as instruction. And this goes back to my point of what's the purpose of education? If it's about purely knowledge transfer, we can see why a lot of ed tech companies are saying we can replace a teacher because all we do is we're going to deliver content and we're going to tell you how it works and we will give you a better description than that product over there. But if you are moving towards what should good teaching, the social, the cultural, the emphasis, the creativity, the curiosity, purposes of education are about. The question of the idea of replacing a teacher with an AI is I think, nonsense. And you look at ingenuity in the States, in Covid, where you can see across the news, you can read up about that, where, you know, marking was done by Edgenuity and parents that were curious went in and found they could type any old gobbledygook, what they call the word salad. As long as they hit or enough keywords, the child got a really good grade. So it was marking just for points. But how do we get away from this? So what's good look like? How long do we wait for it? There is work going on and the UK government, the Westminster government has got the chartered college looking at what does good look like. And this edtech Evidence Board is being set up. It's been established, but why have we not done it with the Education Endowment Foundation? EEF has tons and tons of evidence about what works. You take for example, visible classroom 2018. There was a study published. Over 8,000 pupils were involved. Teachers were involved. It was together with the Behavioral Insights Unit, otherwise known as the Nudge unit in the uk, nesta, AI Media and the University of Melbourne. It was proven by their research with over nearly 8,000 pupils that either it had no effect or maybe even set the children back by a month in their reading. So what does good look like? You know, we know what we're trying to stop. We have data protection laws that we're not putting in place. We don't have the mechanisms for doing that and we need, for example, even just to go back to some basics like what does teacher training look like? We need to have basic data protection, privacy.
- [69:51]
A
I'm going to pass some of these questions to Julian, who is the professor of education here. And I think it does come back, you know, we think we're talking about ed tech, but very often many of the questions come back to education teaching.
- [70:06]
D
So the at the heart behind all of these questions in a way and a lot of them, some of the questions from the floor and actually your question about what does good at any like is really a sort of an unsaid problem which is how can we ensure that there are enough good teachers to go round? How can we ensure that every child gets the best kind of education? So there are real questions of economic questions that are really governing the answer, the governing the growth of the edtech to use Reece's favorite word, to offer solutions. Because what you've got is a situation where as Sandra explained, you have some perfectly good lessons or good experiences with some bad edtech and vice versa and you have. We would no longer. It's just like asking what does a good teacher look like? It doesn't really matter what a good teacher looks like, it's how many of them are there. Now the problem with the edtech.
- [71:04]
C
The.
- [71:04]
D
Choices we have to some degree is are there is the question about edtech really a kind of teacher replacement kind of problem and is it filling a hole in teacher kind of recruitment, retention and above all training in that kind of sense. So it's not by definition these experiences are going to. Every child's experience of education is varied and will have good moments and bad moments. Just like you have some good, you know, from your point of view a good teacher to school and A not so good year every now and again that's kind of to be expected. The question is what is why is the focus on providing solutions to the deficits and problems in standards that we can't unacceptable kind of standards being provided and how does that threat then effectively drive down what's going on elsewhere?
- [71:59]
A
Thank you. I'm going to turn to the online questions because I get a signal that there are some. So thank you. And we're all taking notes.
- [72:12]
I
A question from Nicola Hawkins. What are some of the prerequisites for edtech to narrow the socio economic development provides entertainment and engagement Question for Sandra. What is the impact of defragmented ecosystem EdTech apps on the children's experience? So the instructional time between screen based education and traditional non digital materials. Pen and paper for both schoolwork and homework. A question for Reese. Can you speak to the EdTech market category and where it might be in the hype circle Hype cycle, particularly related to learning tech? And who decided that edtech is inevitable and how could we tell that the lessons Learned from the 2023 UNESCO report and EdTech tragedy were implemented? There's a few more, but I think that's that enough for now.
- [73:10]
A
And you directed the first question to one of the panelists, I think.
- [73:14]
I
Sandra.
- [73:15]
B
Sandra.
- [73:15]
A
Sandra. Yes. Do we see potential for edtech to narrow socioeconomic inequalities or perhaps increase given the incredible variation we saw across schools?
- [73:34]
C
I don't know. Like the first thing that comes to mind is more of an increase than a decrease at the moment. And there's so much variety in the quality of so it's not only what it is about what edtech you have, but also there's so much variety across different schools. What children can access, what they can do at home. I mean the question about was schoolwork and homework. A lot of children don't have devices to do their homework on and that homework is device based. Like you know, it has to be done on the device. They sometimes get the option of doing it on paper. Not always. They can't always do the homework when they want to do it. They sometimes have to wait for if it's a shared device, they have to wait for a parent to get off it. I think there are big inequalities and, and they're growing. Even the example that I gave about different so using a free app versus the industry standard app. What's this doing to your learning? What's the difference in quality of learning that children are getting and the different opportunities they're getting? And I think at the moment the divide is widening and I don't know if anyone else is.
- [74:52]
D
I just support what you're saying, that the research actually shows that the range of edtech is stratified by social class and income, both at school and at home. So it's not just questions of access, it's also questions of variety. Support. So in actual fact, you could, for all the promise of edtech offering a kind of equal learning environment in a classroom, in actual fact, its practical implementation has only been to exacerbate inequalities and injustice.
- [75:26]
A
Rhys, can we get away from the hype, I think was the question, can we bring it down? Can we not overwhelm schools with the marketing where the brand wins over the quality?
- [75:38]
E
Yeah, I mean, I think, you know, I think the hype cycle isn't. I mean, yeah, I don't think it's driven by investors. Is my.
- [75:50]
A
Is my view someone else?
- [75:52]
E
Yeah, because I think there's just so many. There's so many solutions out there that claim to do the same thing. Like I said at the start, you know, how is. How is the school supposed to decide between two things that both claim to save five hours versus six hours? Like, how do you verify that five hours versus six hours is saved? Like, how if it's not, if it's not about learning outcomes and linked to specific sort of verifiable results, how do you then prove that that's the case? And so without standardization, you will always have hype because it encourages marketing. Hype and marketing messages that make you buy the thing as opposed to being. Forcing you to prove that what you're doing works. But the two areas that we see a lot of activity in terms of coming to us, we see a ton of companies that are AI tutors, which I'll go into, why that's a good and bad thing. Well, that won't take too long. But AI tutors, we see a lot of those both selling into schools for use in the classroom, but also obviously selling into parents that want to give their child every advantage to progress. And the other thing that we see is a lot of AI marking solutions which, as discussed, have the same kind of keyword issues where, you know, you add in a couple of phrases and then it gives you. It spits out the mark that you think you've earned. I mean, we. We have backed a company that does a very good job of that, which is better than that. So, you know, it can be done. One more thing, which is that I think. I think the Hype will gradually die down quite soon on the basis that I think if you look across Europe, you look at companies like Arbor Education in the uk, which is a school operating system, you look at Stewie, the company that I mentioned, I think it's possible to argue that 90% of what happens in a school day to day function is the same. And so I think if you look at the suite of tools that schools are going to have, they're going to have a version of basically the same stuff across Europe. And so like how much further can you go away go with that if you assume that's true and you've got teachers at the front of the classroom as either coaches, mentors or as teachers, you've got people that lead the school, that are making operational decisions and hiring decisions, etc. Like those things are fundamentally true of schools more or less everywhere. So I think the hype will go.
- [78:04]
A
I'm reminded of one of the teachers that we interviewed in one of the schools where who had got all the marking done by AI and left school at a sensible hour at the end of the day and went and had an evening and commenting on the other teachers who were still staggering across the car park with their piles of exercise books that was going to wreck their evening but give real feedback. And I could see the teachers dilemma. I mean it is very. It's not like we want them to be continuing necessarily with the current workload. So it's a struggle. I think there was a question about inevitability which I'm passing to you just. It does feel inevitable but perhaps the way in which it's shaped. I don't know that we're going to have tech free schools and roll back on AI and I don't even know if that is what good would look like. But there's something that feels like this is a juggernaut that we can barely stop. Julian looks like he's going to come and say something on that as well while Jenny's collecting her thoughts.
- [79:08]
B
You know, we have choices, we have made choices and we've also had absence of choices. So a lot of things that have come in have been made in a vacuum because schools were left. They're being pitched by companies. There are definitely companies. Some companies operate like pharma companies, they go to schools and take them out for dinner and you know, pitch them and to give them a really nice time and tell them how lovely their product is. So let's not be under illusion how some of this sales work works. But most majority of you know, teachers really are looking for the best products that worked well and do work well for their children. I think the inevitability is, as you say, is only going to be if we've got the right incentives in place for the right outcomes. And those depend on who you are. So if you're government and you're looking for something provable about how we've addressed our learning outcomes for children, you look at things that can be measured and reporting back. And that's why we have progress 8 and profiling and measurable goals. And we've moved to even having, you know, pupil data joined up with your tax record so that you can measure the total expenditure of how much you cost the DWP or bring into the revenue and measure the total cost of your education across your lifetime. You know, we've got real measurability built into some of this, which I think is going to be really, really hard to change unless you change some of the incentives of the government. But I think if you look at it from what's coming is that there's parental pressure coming. And actually, interestingly, it's not the socioeconomic, economic, poverty, digital divide that's developing. It's the white middle class really genuinely sort of, you know, racially, economically, culturally divide that's growing, partly driven by some of these sort of culture wars issues that is pushing back on tech in classrooms, full stop. And I think that is where we have a great political pressure growing that is going to make some of this really, really untenable for schools. And I think government is going to have to step in to give schools proper quality, measured, tested products through something that the chartered college actually does and doesn't just talk about. They need the standards to be applicable. So the AI stuff is great, that came out recently, but it's impossible for school. Schools and teachers actually manage and understand their IT and teacher training needs to vastly improve to understand some of these technologies they're introducing. We could reduce massive cost. And that's something we probably inevitably could say to the government. We could do with an audit. We could simply. We have all these Google forms flushing around the system with lots of pupil data. Why have we never had one that simply goes to schools and says, what pupils products do you use? How much do you spend on them? And we could have a parent survey and a pupil survey. You can make it completely anonymous. You could do it by school, we could do it by school group, by academy group, however you want to do it, get proper feedback and we could know how much this is costing the taxpayer. And then come with perhaps some questions that could be economically driven and change some of these incentives around the inevitability discussion because they're not their choices. But we have to make ourselves vocal and go and speak to our MPs and our representatives and say this matters to us. People like ourselves, you know, have been lobbying government for this for 10 years to get simple things already in law made practical, like data protection laws made realizable, you know, public sector equality duty made, you know, possible to introduce during procurement laws and processes. These things exist, we're just not using them well. But we need to support, support teachers because they are stuck in the middle. They have underfunded under resourced too much to do. They're being asked to do more and more and more in the classroom and simply giving them another thing to do is not going to help. We could do something very simple if we had national assessments of data protection, impact assessments, for example, we could do reduce the workload that every school using similar products does by having much, much better impact assessment measurement at some point, sort of, even if it's regional or supported level. So there are things we could do that don't have to mean every harm is inevitable or every cost is inevitable. I think we want children to be literate and benefit from these tools that do benefit them, but get rid of some of the costs.
- [83:14]
A
Thank you. I think Julian wanted a word and then I, I'm. Yeah, I have my eye on the clock, but you have time for a just.
- [83:22]
D
I think there are two sorts of breaks on the inevitability. One is the choices that people who can afford to make choices are making. So the kind of affluent private schools in Australia, 40% of the school system is private schools. The kind of range of different uses of ed tech and decisions about it. And that's one area that's stopping me. But secondly, we're living through a very strange moment at the moment which I imagine is occupying Sonia's every day, which is a kind of weird resistance to screen free child about screen free childhoods. I think the grounds for that are problematic. But there's something weirdly contradictory that at the same time as you've got quite a popular movement about reducing children's attention to screens, not that we're actually providing youth clubs or green spaces, that's another problem. But at the same time you've got a form of schooling and education which is becoming more and more, for lack of a better word, and I don't really like a screen intensive So I think there are movements afoot that will hopefully break and it will be inevitable unless, as Jen says, there is a difficult kind of investment and support for education and childhood more generally, which is a broader problem facing our societies. And from that point of view, I wouldn't be incredibly optimistic, I'm afraid so.
- [84:53]
A
We have optimists and pessimists here on the panel and I just want to come back to the child Rights framework because I think actually this debate has beautifully illustrated how we start off thinking we're talking about what children are learning for education. We discover it has a whole lot to do with their privacy and data. We had quite a lot to say about their well being and kind of growing up and flourishing as a whole child, as a social being, in the classroom, as a person who wants to express themselves, be creative, follow their own interests. And so it is not possible to kind of draw these lines and say EdTech is for education and the other rights are not part of it. We really do need that holistic approach. And then, more flippantly, if I may, I think what I also learned is we really do want standards and we want them now. And when we get them, we're going to hate them. So thank you very much to the panel. Thank you for listening. You can subscribe to the LSE Events podcast on your favourite podcast app and help other listeners discover us by leaving a review. Visit lse.ac.ukevents to find out what's on next. We hope you join us at another LSE event soon.