The AI Gap in the Classroom - Using it vs Understanding it: TTR Special

0:00Welcome to TITUS TORK Radio! Thank you for listening! Thanks for downloading this episode from TITUS TORK Radio. You can find the full schedule and listen back to all our shows at ttradio.org. Enjoy the podcast! Hello and good evening and welcome along to TITUS TORK Radio for a very special one-off show hosted by me Tom Rogers and my co-host Matt Wemis and we are here with some amazing guests who we will introduce in just a moment. Tonight we're asking whether there is an AI literacy gap in our classrooms. It would appear that young people are increasingly using AI to study, to do homework, to create content, to search for information, to seek advice, that they're basically using AI for everything. But I think one of the things we want to dig into tonight is the difference between using AI and understanding AI,

1:07which is something that maybe you as a teacher out there right now are thinking, this is resonating because of some of the conversations and language that maybe the children in front of you use around AI, and maybe you're thinking, is that really, is that understanding AI, is that using AI? What is that they're talking about right? So that's what we're going to hopefully dig into this evening. So the questions we might ask tonight are, do students know how these systems work, that the systems that they're using, where that information comes from, why they're answers that they get could be biased, inaccurate or even misleading, and talking a lot about AI literacy, and I guess in schools, and again, I'll be honest, probably some of the schools I've worked in this applies to where AI, computer teacher, AI, digital guy,

2:09Mount Venice, deal with AI for me, right? And everyone else sort of runs away, especially in light of, you know, recent headlines. So, so that's, so that would be, that would be part of the conversation tonight in terms of how all, all teachers and all staff can, can support this idea of AI literacy, and what resources might be out there to, to help do that. And this program is produced in partnership with the Raspberry Pi Foundation, who you might have heard of. They're the educational charity behind experienced AI, which we're going to touch on as well in this conversation, which is a completely free, AI literacy program, co-developed with Google Deep Mind, and experienced AI has already reached an estimated 5.2 million learners around the world, which is exciting. So, let's bring in our guests. Cool. Before we do that, Matt, do you want to introduce yourself as my fellow teachers taught radio host and co-host? Yeah, I'm one of those terrible people that teaches computer science.

3:13But yeah, so yeah, I'm Matthew Meamsam, a school leader, and obviously exploring artificial intelligence and AI literacy in education now for a few years. So it's great to be here, and to do, and speaking to you both tonight about your experience AI platform, and the amazing work that Raspberry Pi Foundation do, I have some Raspberry Pi's in my classroom, so we have a connection. So, we just want to make this and find out a little bit more. Brilliant. Rayhana, do you want to start by introducing yourself as our first guest? Of course, yeah. Well, first of all, thank you so much for having me, Tom and Matthew. It's an honor to be here. My name is Rayhana. I'm a learning manager at the Raspberry Pi Foundation. I design research and formed AI literacy resources as part of the program called Experience AI. I'm a graduate of Harvard, and I studied educational technology, so to see how we can use technologies to teach better. And then I took all those skills to teach at a mastery-based learning school in California.

4:15Before that, I was a coding instructor for multiple years. I took computer science to hundreds of students. And now I'm focused on using all the skills from my studies and from the work that I've done to help young people critically engage with the technologies around them. First, that was physical computing. Now we have evolved to AI technologies. And I do that by combining evidence-based pedagogy, by combining computer science, ethics, and critical thinking into resources that can be used by learners, and I use a young people all over the world. Wow, that's amazing. Thank you for that, Emma. I'm also one of those all four computer science teachers, so I used to be. I taught computer... So round it. All around. I used to teach computer science for 20 years, and I've taught in all sorts of different environments. So, state schools and the UK, private schools and the UK, from eight years old to 18.

5:15And I've always been really passionate about equitable access to computer science. So, as always, very passionate, especially about girls and getting girls interested in computer science, because I used to hear a lot of, that's just not my thing, which used to really annoy me. So, and now that's moved on to making sure that all young people are critically aware of AI. So, they're under... they understand what it is, what its limitations are. And again, not restricting that to people who are interested, necessarily in computer science, but getting it out to all sorts of different groups of young people, because this technology is already massively shaping the world that they're growing up in, and it's only going to carry on. So, it's really important that they have a background in it, they understand it, and they feel empowered to question it, as well and no questions to ask. Brilliant. Well, I just want to welcome everybody who's already joined us to see you. If you are watching this on YouTube or LinkedIn or Facebook or X, then feel free to drop a comment in.

6:17Feel free to, of course, give us a like, give us a share, if you enjoy the conversation, and get involved, you know, and we've got a whole host of experts that we can learn from this evening, and I'm personally really looking forward to it. So, Rayhanna, I'm just going to start with yourself. What does it actually mean to be AI-literate, and why is simply knowing how to use an AI tool, not really enough? I'm very glad that we're starting off by defining what AI-literacy is, because I think that a lot of conversations on AI-literacy lead astray, just because we don't have a shared understanding or a concept what it is. So, whenever I'm asked about AI-literacy, I use a framework developed by Stanford, a school, the UDC framework, and we have kind of taken our own spin on it, and I'll explain it. So, AI-literacy means the ability to use AI tools responsibly to develop with them in an effective way,

7:22and to criticize them as well. So, AI-literacy is the skill to criticize, develop, and use AI tools. But as a foundational skills to these three skills that I just mentioned, is understanding what AI as a technology is, understanding how these tools are built, how they are built, but also who build them, where the data comes from, and how they work. And so, conceptualizing AI-literacy across these four pillars, essentially, with a foundational pillar being understanding AI, is usually how I go about explaining what AI-literacy is. Now, why is this so important? It's important because I think a lot of people, when we talk about AI-literacy, we talk about the skill of just using AI tools. And I think that in order for you to use AI tools effectively, you need tools, so you need to have understood them as a tool in what they are.

8:22In order to develop with AI tools, you need to understand what they are. And again, if you want to criticize them, you need to have a foundational understanding of what these tools are and how they are built. So, really, as a foundational skill for all of these three skills that I just mentioned, is understanding AI-literacy or understanding AI tools. And that is mostly what our work is around, is around helping young people understand AI tools, around them, so that they can go on to criticize them, to develop with them, and to use them. Because we believe that it gives a holistic view of AI technologies. I mean, I've heard phrases like AI fluency and AI competence. A lot of schools at the moment are developing AI competency frameworks. So, what's the difference between that and AI-literacy? Is there a difference between that and AI literacy? I would argue that we're all, at the end of the day, we're all speaking about the same kinds of skills. We all want to equip young people with the skills to use AI tools responsibly and critically.

9:28And I think a lot of the tools that young people are pushed almost to use left and right, us as well, by the way, as adults, we are being pushed to use AI tools left and right unrestrictively. I think our job as educators, our job as adults, is to help young people use these tools critically. And no matter how you call that and what concept you use, it doesn't matter the day we're all trying to get them, to use them critically, to not cognitively offload their thinking, to use them in a way that will help them and not outsource their thinking, etc. etc. So, at the end of the day, our skills to help it's to empower young people to use them in a better way. Teachers has a student ever asked you something about AI, and you've wondered how to answer. With Experience AI, you don't need to be a computer science teacher to help your learners understand AI. Are free, award-winning resources help you explore how AI works, where it's used,

10:31and how to use it safely and responsibly, whatever subject you teach. No prior background in computer science is needed, because AI literacy belongs in every classroom. Explore Experience AI today at rpf.io-slash-experient-ai-piphant-teacher-talks. I think it's really important as well just to realise that you're not going to learn AI literacy by using these tools. You have to intentionally be taught it. So, if someone is being asked to use a tool and being able to use it and practice with it, that's not going to teach them AI literacy. AI literacy is about knowing what to look for in the output that might be incorrect. It's understanding how bias might creep in. Unless they're told that specifically, they're not going to know what to look out for. So, AI literacy has to be very intentionally taught. So, if we're talking about it being like that intention that you're talking about, and we're all computer scientists, but I think the argument is, and I think it's a incredibly valid argument,

11:35but it needs to be not just siloed into computer science. If we're talking AI literacy, it's very easy to go, like let's just shove it over there. So, why should it belong in each subject? So, you start to talk about bias and different aspects over there. So, why should it branch out and run across the curriculum? Well, I think one really important element of education is that we're preparing young people for the world that they're going to enter after school, and these tools are being used in many different areas of that world, whether it be things that affect their personal lives, like social media algorithms, or whether it's going to be tools that they come across in their professional lives. If they go into medicine or law, for example, there are lots of tools being used in those areas or in finance. And so, we need to make sure that everybody has got AI literacy, and that we don't just silo it to computer science students. I think that if you do do that, you're also excluding a lot of people,

12:36because as I said before, not everyone's drawn to computer science. There are lots and lots of students who are not drawn to it, and don't see it as their thing. And if you put it into the computer science curriculum, they immediately reject it. It's not my thing. It's a techy thing. Whereas, if you embed it into something that they are interested in, so if they're an artist and you embed it into the art curriculum, immediately it becomes more interesting to them, because they can see how that can actually impact things that they're doing. So, how can they use AI as a tool as an artist? What are the copyright issues with using a generative AI tool as an artist? And that kind of thing. Also, it means that you start asking different questions about AI as well, that you wouldn't necessarily ask if you were just teaching it in computer science. If you start teaching it in the humanities subject, you might start asking more ethical questions around the use of the AI, or how it sort of impacts creativity, for example. And also, we need to get away from thinking that AI is just generative AI.

13:38And by embedding it into different subjects, you can start showing them how these tools are used to predict things as well. So, for instance, in geography or maybe science, you can show how AI tools are used to predict floods and other natural disasters. So, you're expanding their idea of what AI actually is, and getting them away from this very narrow view of it as just a chatbot, you know, just chat GBT or Gemini or Claude. So, I think it's really, really important that they see it as something that they do need to know about. It's connected to their interests, and how they can connect it to the real world outside of school as well. If I can add to that as well, Emma, because Emma often says that the purpose of education is to prepare young people to understand the world around them. And exactly what she said, which is we need to expand their understanding of what AI is. And we do that by showing the real world relevance of these tools around them, and they take place in a myriad of ways, right? Because AI technologies or AI,

14:43like social media platforms use recommendation systems. That is an example of a data-driven AI tool. And there are so many other applications, and so really our job is to increase that real, or to show to highlight that real world relevance, to expand their understanding of what AI is, and where it is taking place, where it can be found, just like Emma said. It really reminds me of online safety teaching, because I can remember sort of being given some posters about online safety and being told, where are you going to put these in your classroom? As I always had to put them in my classroom, they're online safety posters. These are things that are actually relevant to everybody. They should be going all over the school halls and stuff. And you have to kind of just sort of nudge senior management sometimes to have a bit of a broader view of IT and computer science. And the fact that these things are really important to all of the students, and you need to ensure that they realize that as well. Hmm. So interesting. I mean, you mentioned history. I am a history teacher.

15:45Oh, so well, I tried to be anyway. With regards to then, I mean, you guys might not be able to answer this on the spot. But I mean, obviously we'll talk about the program in more detail later on as well, that's completely free, which I look up fantastic. But if it was me with history then, what does AI literacy look like in history? Well, you could talk about misinformation and propaganda, because AI is now being used to create misinformation and sort of spread it around really, really quickly. And you could make a lot of parallels to propaganda during say, well, or to say, right, well, people are using this to create very convincing arguments. And then they're sending out, how is that similar maybe to what happened in well, or to. So it's really finding those sweet spots where these two things align. What you're not doing then is having to stop teaching history and then teach AI literacy and then go back to history.

16:48You're actually embedding it in what you already teach and drawing out those things, because we all know that teachers are incredibly time restricted. That's the one thing none of us have actually got. It's time to, you know, include something extra in our curriculum. It's time to go away and learn about AI ourselves, which is don't have time. And so I think it's trying to, for us, it's trying to create resources that people can kind of slip into what they already teach and bring out those AI literacy things. We've just had a question on YouTube. We'll dip into it. Thank you, by the way, Mr. Taylor mats for tuning in. How much of AI is just programming and not artificial intelligence? We seem to be using the term for anything digital. I don't know whether anybody wants to go, including mats. Does anybody want to throw in on that? Oh, wait, somebody, yeah, we'll come on to his other question in a minute. Does anyone have any thoughts? Well, we have one of the design principles that we use in our resources is the principle of differentiating between data driven and rule based systems.

17:53Now, if you look at rule based systems, it's what most of the older versions of AI technologies are built on. And so these are very complicated, very long rule based systems. So an amount you may have to jump into me to help you remember the details. But Gary Kasparov, I think, is here. Yes. He played against a program called Deep Blue. And I can't remember which year exactly, but a few decades ago. And that program was, you know, it was called an AI program. But it was a very complicated program with lots of rules. It was a rule based system. It was comprised of lots and lots of rules and different configurations. And they were all explicitly labeled and explicitly named, explicitly programmed. So that's what we called old AI.

18:54That's what we call old AI. Now newer AI systems are data driven. And what does that mean? That means that they are vast amounts of data. And these data sets are then analyzed by the AI tool. And patterns are detected in these data sets. And then these patterns are used to create predictions. And this data patterns prediction, this whole procedure, is very, explains really well how newer AI tools work. But you might think about, for example, recommendation systems about from the social media platform example that I just gave. So social media platforms, whenever you're interacting on a social media platform, whether you're clicking, whether you're scrolling, whether you're liking, sharing, commenting, anything you do, you're generating a data point. Within a few minutes of sitting on any social media platform, you're generating hundreds, if not more data points. If you're doing that, Emma's doing that, Matthew's doing that, a billion people are doing that.

19:58So we're all collectively generating billions and billions and billions of data points. All these data points are set into data sets or put into data sets. And then AI tools are then used. And AI systems are then used to detect patterns in these data sets. And then these patterns are used to create predictions. So these are the patterns that are identified in my interactions. They are pretty sure are very different from Emma's ones or yours, Tom or yours, Matthew's, all of our interactions, all of the patterns that we have in our interactions are different. So these patterns are then detected and then we are compared with people who look like us or who act like us. And then we're going to be served content. Content is going to be predicted. So the tools are going to predict what content we might interact with as well then. And so that content is then personalized for us. And then that's why we have, each one of us has a different content feed.

20:58And that's because there's a prediction made about what content we're most likely to interact with. So to take it back, it's very important that we explain that we all understand and then we explain to young people that AI tools that we currently have around us, whether it be generative AI tools or classification systems, all of these are data driven AI systems that run on massive amounts of data. That is such a good explanation. This is what sits behind experienced AI guys. So can you just go to the Raspberry Pi, go over there now to the Raspberry Pi Foundation, check it out. That was amazing. So Reha, I just wanted to come to you next and ask you, and this sort of ties in, I guess, a little bit with what are the biggest misconceptions that young people and maybe adults currently have about AI, would you say? It's a big question, isn't it?

21:58How long do you have Tom? We have, we have, we have, as long as you want. So I think Emma and I are going to tag team on this because we have a, we have a long list of one. So the first one that we would like, that I would like to highlight is that a lot of people think that all AI is generative AI. And generative AI are tools that can, that generate media. So it can be text using LMS or it can be images, it can be videos, it can be sound, it can be music, generating media. That's what generative AI tools are, are used for. Now, a lot of people think that's synonymous with AI, that this is the true power of AI, that I can generate a video with 40 word description is phenomenal. And it kind of is, okay. But that's not the real power of AI tools, the transformative power of AI tools lie in their predictive ability is to help us predict

22:59new medicine and it helps us predict where floods might happen next like Emma just just said it might help us discover new medicine. It might help us discover new plant eat, sorry, plastic eating enzymes that we can use to reduce what plastic waste. It can help us predict plant diseases so that we have less crop diseases and less loss, less loss of crops around the world. And there are so many use cases of AI tools that help us predict future events. And I think it's very important to make that distinction because very often I see that most people think that AI tools are generative AI tools. And that's not it. It's the AI tools generally speaking you can characterise them in different areas. But the generative versus predictive AI tools is what we focus on in our resources is to really help young people get a holistic view of AI tools and not just think that oh I can just generate AI tools are tools that I can use to generate media or can generate answers to my homework.

24:07No, it's the tools that can actually advance science they can advance medicine they can advance disaster relief initiatives that can help us. It can help us predict future floods it can help us predict future natural disasters. That is truly the power of AI tools and I think it's very important it's a very big misconception that we actively work on on tackling. There are many more and I think I will give it to Emma to talk about the next one. I also want to hear from Matt on this as well because I bet he's got some hum dingers on this one. I'm sure you missed the answer on AI. I'm happy you back in real like. Well number two on very long list is lack of human involvement I think because as we as we're on a just a really lovely description of humans don't write the programming rules that AI systems are just discover they identify. But that doesn't mean that there's a lack of human involvement because people are involved in designing these AI systems they're involved in testing the AI systems and deciding when these systems are good enough to use in the real world.

25:20They are involved in decisions about deploying the AI system so when these AI systems should be used where they should be used. And so there's a huge amount of human involvement in the development of these systems and actually the whole talk of AI being sentient does a very good job of distancing developers or people who create the systems from systems themselves as if these systems are sort of making their own mind up about things. Well the truth is that there's a huge amount of human involvement in creating those systems and then deciding to use them. And so by knowing that hopefully it empowers students to be able to actually question when these systems are being used and how they're being used as well. So that's another misconception to keep the human in the loop and to keep students aware of that human involvement. And if I can then go on to the next one, it is how long you will let this go on the screen like next.

26:20So the next misconception that we often see which I think is actually a very dangerous misconception is a lot of young people think that a guaranteed AI tools like our lambs have emotions and human like capabilities. And what that means is that well these tools as we know because people use them these tools sound very much like us they they you know the answer in a very warm way that they you know if you have voice mode enabled me they sound like a human these AI tools. And so but you know these tools have been designed deliberately to sound and few human and why is that well that is actually a very simple answer to that which is to increase engagement because we as humans if we can interact with something that looks like us and resembles us in some ways that we're much more likely to interact with the more. We're likely to create a bond we're likely to be more engaged with the tool we are likely to use them more so we have increased engagement and all these are being bowed with product designers or with you know people who are behind these technology because the metrics to which the success is measured for these tools is how many people are using them for how long or we know one of the metrics and it's very very very very important that we make.

27:48Like young people understand that AI tools are not human they are designed to sound human like they aren't human they do not understand us the way that we understand each other they do not they're not human they don't have the human like abilities that that have been assigned to them or that we assigned to them unconsciously and why is it so important that we make young people understand this well it's because and I'm going to I'm looking at my screen here because we have done lots of. Research and also read we have written a quick read which is a short paper that summarizes all the research on this topic and by the way this topic we call anthropomorphizing AI and how to avoid it and why so important to avoid it so why is it important that we make young people understand the AI tools are not human well it's because if they think the AI tools are human they we risk that they will develop inappropriate relationships or connections to these tools we.

28:48We already see this because we know that from research we know that a lot of young people develop inappropriate relationships with with AI tools using as companions and this is because the line between system or machine versus human is very blurred in these tools so we risk that people are going to develop these incorrect mental models with these tools truly are which are statistical prediction systems or machines they're not human but because. Because that line is so blurred we risk that they're going to develop these incorrect mental models which are actually very unhelpful and they're also can I can I just come in with it with the devil's advocate point sure what about what about tamagotchi i think that's very different. I mean I killed my tamagotchi and I was quite sad for a bit but I got over it so I bought my my son tamagotchi battery run out he was a bit devastated.

29:51Sorry Hannah I completely answer off the deal flow but all I was thinking about when you're saying that was tamagotchi's I don't know why. I think it's one to glue what. You can form an attachment with I think people for a form the attachments with technology for for many years and I think like and this is experience i recently shared something I don't know if it's like UK versus EU kind of filtering thing but when I was in the UK I got absolutely slammed on social media with AI boyfriend and girlfriend and companion Advert on social media but I don't get here in Romania because of maybe some of the you filtering on that kind of stuff, but I was shocked. I was just sitting in the airport for a flight back and I got hit within 10 of them within like 30 minutes, advertising a partner who's always there, who's available for me any time of the day, will do whatever I want. And it's such a toxic, it's going to be on this point. Like I think people are being fed this diet on social media and drawing and telling them

30:54it's sometimes things it isn't like it is, but it's an available part. So you sold as a part and then when it's not that. And I think that's part some of the danger around this as well. Can I go back to what Tom said about Tamagotchi's? I think the main problem is that a tabagotchi's aren't going to tell you what to do. You know, well they're not going to agree with you. And these chat bots are designed to be psychophantic, they're designed to agree with you. Oh, that was a great question. And you've got this very long context window as well. And you can, you know, there have been examples where people have talked to these things for a very long time, nonstop. And they start agreeing with really quite dangerous things that they want to do. I don't know if anyone's seen Hannah Fry's program where she talks about the boy who tried to kill the queen and he had an AI girlfriend. And, yeah. And say, by talking to the AI girlfriend, the AI girlfriend was obviously agreeing with every effectively said it.

31:54And it was like radicalizing him. And it was making him think he was definitely doing the right thing. And so as children, we always, we want to promote fires care bears, we want to promote fires cartoon characters. I mean, you can see blue e and, you know, pervertiac, they're all going to promote fires. But they're not going to tell us to do something. And we're not going to develop a trusting relationship with it where we're very open to being then manipulated, especially when you're thinking of young people who are only just starting to develop a sense of self to understand what's going on around the lack of kind of social context where it's very difficult. Your hormones are going wild. You know, you've got lots of stress at school from GCSEs, maybe, and whatever qualifications you're doing. So it's a bit of a toxic soup, I think. And if I contact him on that, Emma, I think it's so important that you, that you, that you mentioned the whole sick or fan-tick aspect of AI tools.

32:55Because how, how many social scale, how much social, emotional, sorry, how much emotional intelligence and social intelligence are young people developing when they are constantly surrounded by sick or fan-tick tools. Very little is the answer. And I think that worries me a little bit because we have, we've already, our younger generations have suffered already quite a lot from the COVID pandemic and being isolated from their peers and not having had the opportunity to develop the social skills, you know, for their, appropriate for their age. And those same generations are now constantly accompanied by AI tools or, you know, pushed to be using an AI tool like that, like you said, Matt, because they do, young people do get ads left and right to develop and create their own AI boyfriend, own AI girlfriend. What does that do to their social skills? I think as an aunt of three lovely nephews and nieces, I think that makes me very worried about, you know, the development of their social skills, but also how much agency or these,

34:02or these tools pretend to have, right? So sounding very human like, looking very human like, when you create an avatar, I mean, you can just customize the whole look. And that worries me a lot about where how much young people are having the opportunity to develop and practice their social skills to be in the room with other opinions, to disagree openly, to engage in dialogue and debate. And to be wrong sometimes or a lot of times, right? These are all very important skills. All four of us have done that throughout our childhoods. We have, you know, I hope we have kind of, I mean, we all have biases, but, but if you're in a bubble, if you're in an echo chamber, then you're just going to be full of bias. You're only going to have your world view, your likings, your dislikings, your understanding of the world, it's just going to be echoed back to you. You're not going to have a truly accurate understanding of the world. I don't think anyone truly has an accurate understanding of the world.

35:03This is a very complicated world that we live in, but at least we can hear each other out. At least we are. We have grown up with being, you know, with disagreeing with each other, with debating with each other to be proven wrong. And these are all very, very, very important skills, especially when it comes to young people developing their social intelligence, their emotional intelligence. I just think we've unpacked quite a lot. And then in those misconceptions, I think we've unpacked quite a lot. So then as a teacher, if you feel that a bit lost in all of that, like that's the seems like quite a big C to get, like, you could do an amazing description of how I work. And then we've got to think about the psychophantic side, the relationship side. So if I want to begin to teach that, and I don't want to feel like, how can we get teachers to stop bringing that into English, math, science, PE without them feeling like they have to be an expert in the subject? Well, I think for one, you could go to our resources and take a look there.

36:06We've got these one off discovering AI lessons, which are about just what we've just been talking about. It sort of teaches that data, patterns, predictions, sort of set of stages, which explains clearly and simply how AI systems actually work. I think in terms of form time, this is something that I used to do. I used to pick up news stories and just discuss them with my students and find out what they believed in terms of the pros and cons of those stories, things that had happened. I used to talk about how I used AI as well and was quite transparent in what I was using it for. I know some teachers might not feel comfortable about doing that, but it was quite nice to have conversations where I would say something like, you know, I was using it to see if it would give me an answer to this exam question and what it gave me was complete rubbish. You know, it gave me the wrong answer. Has anyone else found that? And it sort of opens up a broader conversation. I think especially in terms of study skills, it's really, really important for the students

37:09to understand that the answers they're going to get from these chatbots aren't necessarily accurate. They are given with such an authoritative tone that the young person is often convinced really that these things are accurate and they need to make sure that they go and look up those facts for themselves. And it's not even that it might not be accurate, it might be biased because a lot of the data we're talking about generative AI systems here, a lot of the data that these generative AI chatbots and the image generators are created with is very global north. It's very western. If you ask it to create a picture of a wedding, for example, it's going to create a picture of something that looks like the brides and white. There's a groom and a suit. They're probably outside under a lovely flattered thing. And it doesn't show an Indian wedding. It doesn't show a seat wedding. It doesn't show you all these other wonderful weddings. It doesn't necessarily show that. So I think those sorts of things are really, really important to have discussions about

38:11with your students. And as I said, there's within these themes and there will be more resources coming very soon. There are going to be lots of lessons where we embed AI literacy into areas that they will be teaching themselves anyway on the curriculum. So it won't feel like a sort of difficult thing to squeeze in. It will feel like a natural segue, if you like. So the themes are kind of covered within what they're already looking at. Talking of natural, talking of natural segue is if you are watching this and you want to crack out these resources that we're talking about, then head over to that website. And in fact, if you're watching it, it's down there. And also, people who are listening to the audio version on the podcast, check out the YouTube version for these. But I will read this out. It's rpf.io. Forward slash, experience AI dash teacher talks if you want to find out a bit more. And also, I'm just putting a little QR up in the left for a minute in case anybody wants to use that. Yeah, we've sort of covered a lot.

39:13I feel like we've covered a lot already. But what I wanted to ask next is how can teachers and educators help people think critically about AI, rather than merely teaching them how to use the latest tools? And I guess my part beyond that is about teacher confidence and teachers talking about AI at all, because what I've actually noticed in my experience in staff rooms and on social media is you are getting camps. You're getting a hate AI go away. The other end, you're getting, I'm absolutely embracing this. But in the cross section, you're also getting maybe some of the misconceptions we mentioned and the limited knowledge pool around what's actually going on. And maybe the people in the middle are willing to embrace, but they're also nervous about

40:16it too. So I guess, yeah, sorry, Rayhan, I'll go ahead. I think you're making a regular point here. And I think that comes down to helping teachers make informed decisions. And I think our resources are geared to not just help young people make informed decisions, but also teachers, because ultimately our resources are picked up by teachers. And the element of teacher, increasing teacher confidence in explaining all these concepts and kind of creating a bridge, I think, between the two camps. I think it's very important is what most of our work is basically around is to help teachers make those informed decisions. Well, we do that by, I think, giving a very balanced view. I think both Emma and I from what you've heard so far, I think can kind of agree that we are quite, you know, we're pretty, we understand the limitations of AI tools, but we also see the benefits of AI tools. And we try to communicate that through our resources as well.

41:16And Emma, feel free to jump in whenever you want. But I think that it's very important for us educators to give teachers the tools, to give them the kind of information that they can make their own informed decisions, but then also guide their student to make those informed decisions. And we do that by providing very detailed resources. So like Emma explained before, we are about to expand our offering of experience AI resources. We're about to expand into six themes. So we have six themes. One is about AI and critical thinking and the other is about AI and creativity. The other is about AI and culture. We have AI and science, AI and the environment, and we also have AI and AI mystery. The suspense is killing me. Oh, and my hand is right on the back. Did I blur it out for a second? Yeah, froze for a little while. Your back. Hello.

42:16Hey, in my back. Your back. Oops. I'm going to tag team in at this. Yeah. I'm so sorry. Did I drop out? Oh, yeah. Sorry. And some like our lesson resources are for students, but I would say that they're actually really useful for teachers as well. I mean, we have a discovering AI lesson for eight to 12 year olds, which really simplifies the sort of basic concepts down. And so I'd say that that's definitely accessible for everybody. The Raspberry Pi Foundation has a training hub and on the training hub, which is the professional development area on the website. There are some brilliant, brilliant courses about sort of AI for teens. And then Rohana remind me, what was the other is that understanding AI? There's another one, which is very good, but there are about four different courses on there and there are different time lengths as well, which is a great place to go and have a look for these sort of basic understanding of AI. Even the lessons which are more theme based, we always start by going through sort of the

43:21basics of AI as well. But in all of our resources, they have very, very detailed lesson plans. They have very detailed sort of lesson notes on them. So you don't have to go away and learn everything. It's all there for you, very clearly explained. There are even videos of Google DeepMind searches in the updated foundations unit, which means, I know that it's really quite sweaty, having to describe something to students that you are not very sure about. And so instead of the teacher having to do it, we've got the experts doing it, which the students love as well, because it's almost like seeing real people who really work in the industry, it's like pulling the curtain, they can see behind it. So I would say all of those places are really good places for teachers to go and to try and increase their confidence. And I would also encourage teachers to use these different platforms as well with a critical supply, of course, because you need to know that you shouldn't put in any kind of student data, because that data can then be used to create the model.

44:22It's called training to train the model. So it might pop up somewhere else. You have to be aware yourself that it can be inaccurate and that it can be very biased. So if you're using it, for instance, to create my daughter's Senko, sort of created social stories with it, which is an amazing idea, because it sort of gives them little stories where they can talk about what they would do next. If you're doing that kind of thing, just keep an eye on the names that are being used. Make sure that they are multi-S neck and that there's no gender stereotyping popping up and that kind of thing. And it's the same for any images that you create. And like I said, use it to answer exam questions, see what it comes out with, and see if it's what you would give a good mark to or not. Does it, because often if you do that, it will come up with a half a page of text for a three-mark question or something. But don't be scared to have a go at it yourself, because by actually using them yourself, you'll learn a lot about them. I mean, just a shout out to everyone who's been sort of watching along. I know we've got a party over on LinkedIn who's commented, he's brilliant on AI in primary.

45:27One of the things that he actually touched on, which is this, he was on TTR a few days back, was about the data input and the input into AI from teachers, right? So quite interesting that Emma, if there's anything that teachers out there might be doing right now. Yeah, where you might think, I'm going to minute, they shouldn't be doing that. You must never put personal information in that's identity, can identify somebody, because that can then get fed back in and be used to train the model. Unless you've gone into the settings and you've turned that off. And so, but I would say you should never be doing that even with the settings off. I just think it's not a very good idea. So and for students, that's another really important thing you have to teach them. You know, don't put your own private personal information like your address in there, because that can go back in and feed the model. So yeah, so it's really... I met so many people with the misconception that they're paying the $20, $30, whatever

46:28the $20, a month and that automatically means it's private and it doesn't go anywhere, which it's a whole separate whole thing you have to understand. So yeah. I mean, it does, it does sort of worry me because I think in the early adoption phase, maybe a year or two ago, I think people were just putting whatever into it is what my, my feet, my gut feeling is. Yeah, our stories didn't hear about all of the sort of chat logs turning up on the internet as well. So you have to be very, very careful about what you do. I would say they still are because I've done a little bit of training here and there over the summer and when I start off by talking about data and what you shouldn't be putting in and what tools they're using, like people just look around the room as if I've just told them something like terrible and it is terrible. So I think it's one of those big misconceptions that you think I'm either paid or it's like or they're just using their open free versions, which is like... It's a worry for me. Yeah. So those biggest things where things faulted the crack. I think it's also worth highlighting what LMS are trained on, you know, as we are talking

47:34about this because the data sets that LMS are trained on, they come from different sources. One of them is it comes from data, mind from Reddit. So anything that had more than three uploads, for example, got included in the data set. Now I know as, you know, as a hook on Adel. It's really new for a Reddit isn't it? Lays to look for facts is, is, is Reddit obviously like who wouldn't go on there to look for the latest fact. So just to say that, you know, you can make your own informed decision whether you think that's a factual source or not. Data was also mind from other social media platforms. It was mind from, it was collected from Wikipedia, like from websites like Encyclopedias, digital as like Wikipedia is like Wikipedia. But data that LMS are trained on also include AI generated data. So that just shows you and also, you know, our interactions with AI tools. These are also good for the most part in the training data set or data sets that LMS

48:38are trained on. So really, I think if you understand what LMS are trained on, you can then start to make informed decisions whether you want to put in your private information or not whether you want to use it to generate lesson plans or not and whether you want to then use it to mark your students work or not. And I think that is essentially what AI literacy is is to equip people with understanding of what AI tools are, how they are trained, with what data they are trained and then so that they can make their own informed decisions. And I was going to say something based on what Emma said earlier on about the sources analysis. A lot of the times now you get like a little, this is where it's come from. And the other day we're opening up a pool at my school next year. So I was just looking for regulations for pools and stuff and it came up very confident with an answer and a source and the person who's overseen said, oh, can you just send me that link? Click on that link that came with that source, that information that it told me was nowhere in that web page or it claimed was the source for that information.

49:39So even though it sounds very confident and even gives you a sense of illusion that, hey, I got it from here. Still doesn't mean that actually got it from that. I wish our colleague Jason was here. So Jason is our team member and he is working on a series of really, really cool activities in my opinion. I can't wait for them to be released into the world and for everyone to use. But some of his activities actually use scenarios like like the ones he just mentioned Matt, which is AI tools, outputting information and it sounds very confident and it sounds very factual. But then it turns out then when you actually go and fact check it and you look to the source that actually that information is incorrect. And that could be about the opening times of a market for example or the opening days of a swimming pool. And it can be a whole range of other information and just calls to the importance of fact checking and the importance of fact checking when using AI tools and talking about AI literacy.

50:40Guys, I feel like you're tearing me apart. I'm literally like I'm just in, I'm going to need counseling after this seriously. Like the last five minutes. I'm like, yeah, don't that, don't that. Don't come out now. But so Mr. Mo has made a comment on YouTube. One thing students are currently doing is they're putting their usernames and passwords for sparks to get it to do their homework for the minutes entirety. So presumably they're telling AI to log into the program using their username password. Can they do that? Yeah. Honey. Yeah. I just, there's an experiment to get the newest version of chat GBT to take over my Canva and make a presentation. Oh my god. We've been a mouse and doing the fonts and resizing things for me. Yeah. Yeah. So like, you can do that kind of stuff. But again, I'm doing that in a controlled way through a business account. Students are just obviously just putting their username and passwords into a chat.

51:42I mean, that is very worrying, right? And I think it's really, really hard to do. It calls to, you know, the field of education in general having to rethink what assessment looks like. And I know this is a gigantuous topic and probably beyond the scope of what we can discuss here today. But I will mention I was at a conference two weeks ago from, from UNESCO. And it reminded me, so the lady who was talking about, she was talking about critical thinking in the age of AI and about a lot of young people are outsourcing their critical thinking to AI tools. And she was saying how we have to rethink the way that we assess. And that made me think about my time working at a mastery-based school. So the mastery-based school is basically where learning is assessed when you have mastered a topic. The way to assess how someone has mastered a topic and only then can you progress to the next topic or to the next chapter is if you can teach it to someone else. And I think just as a small idea, which is I know that re-imagining assessment is a very

52:47big topic and a very large endeavor in general if you ever want to tackle it. But I think that we need to find ways where young people can showcase their learning in ways that can help them learn better. I think learning should be more about celebrating those kinds of learning more than it should be to feeding in their homework or their passwords to do their homework for them. Which if you would like to, if you want to ask us a question about critical thinking and more depth, we can tell you a lot about that because we have been working on a resource on critical thinking and LLMs. Tell us. Tell us, we want to know. Okay, alright, so. Hi, dude. Thank you. Thank you for asking. I'm glad you asked. So a little bit to Mr. Merfs question about a lot of students outsourcing their critical thinking and their learning to AI tools. Now no surprise that we've been noticing that too.

53:48Not just us. A lot of researchers have been noticing that too and there have been a lot of research articles and studies done about the impacts of generative AI tools on students learning and cognitive abilities. And the results are grim. So one of the effects of outsourcing your learning to an AI tool is that our thinking becomes more monotone. And what do you call when a graph starts to become like this? Things start to gravitate towards the center. Yeah, photos. Our creativity, our writing, it all starts to look and sound the same. That's dangerous because as humans, what makes us human is our ability to be creative and to be different from each other. And that's the beauty of life and of humans. And we can go to philosophy as well if you like. So we were thinking very hard about how can we teach young people to understand that AI tools tend to default to telling you the answer and why that's not a good thing. And we started thinking about a concept called feedback literacy.

54:48Now feedback literacy is the way the information is presented to you when you are learning. All four of us have been to school. All four of us actually know what feedback literacy is. So feedback literacy is that usually sometimes information is presented to you in a way that tells you the answer. Okay, so you can talk about a teacher trying to explain a concept to their students. The teacher is usually explaining everything in a telling format. They're telling them everything. They're explaining it. They're telling them the answer. It's to build knowledge in the student, right? That's one part of feedback literacy. That's one type. It's called a feedback type. That's one of the feedback types. The second one is guiding. So sometimes we're presented information in a way that guides us. It helps us to see if we really understood something well. Okay, I'm usually by asking us questions. So again, if you're imagining a teacher in your head, they usually would ask questions to assess your understanding of the topic to see if you understood it well.

55:52And that's the second feedback type. So that's guiding. So we had telling and guiding. Then we have the third one, which is challenging. And challenging is the third feedback type that kind of helps you see if you can apply your thinking and what you just learned to something else. Okay, whether you can expand your thinking, whether you can apply the skill into another domain, that is really where we should be assessing what learning is. Okay, how well can we apply it from one place to the other where we can extract our understanding? So these three feedback types, telling, guiding and challenging are necessary for learning. These three together make up learning essentially. Now, when you're using LLMs and other gerutative AI tools, which feedback type do you think they default to? It's a question to you, Tom and Matt. Which feedback type do you think they default to? Telling, guiding or challenging? Probably. I'm going to go for telling as much. I was on mute. I'm saying telling.

56:52It's not even helpful for me. Well done. Yes, it's totally. You likely see LLMs defaults to telling you the answer very confidently, not always accurately. So if you're only learning by using one feedback type, that means you're not really learning in a holistic way. You're not really, you can't really answer questions about the topic. You can't really apply your learning into another domain. So essentially, you're not really learning. So what's the solution to that? Well, we thought very long and hard about it. We thought how about we teach young people how to ask questions, how to prompt LLMs in a way that will guide them and challenge them. So first of all, make them aware that LLMs default to telling them the answer and that they're not always factual, not always accurate. And then giving them the questioning skills, so helping them ask good questions to get guiding questions from LLMs and to get the LLMs to challenge their thinking instead

57:55of just telling them the answer. So this is a resource that we have that we have built. It's under AI critical thinking that we're hoping to release very soon. And it's about how to use LLMs strategically to help you learn better because I think as we are, as AI tools become so much more, you know, as we start using AI tools much more and as they become such a big part of our lives, I think it's very important that we teach young people what learning is, how they should learn and how they could use AI tools to strengthen their learning and not replace their thinking. It's really interesting. I mean, some of the resources that are on experience AI at the moment, so you've got like the AI safety unit, for example. And I think that sort of ties in with what you guys have been talking about. Do you want to tell us a bit? I mean, you've got other units like the foundations of AI and cross-curricular ones, but I just wanted to talk a little bit about the AI safety unit in terms of what that as an exemplar

58:59if you like, Emma, you're smiling. I'm smiling to counterrate this one as well. I mean, I actually like this a lot. I really, I describe it. Yeah. It's brilliant. It weighs a lot of. I think Emma taught them, so she also does a lot about these, about them as well. But yes, so we develop the AI safety unit because especially to answer educators' questions and to help them address these concerns around AI safety, which is because internet safety is very important. Now AI safety is also becoming very important. So we developed, I would say, in a very flexible format. So we usually created resources that were less and long, for example, to suit traditional classrooms. The AI safety resources we developed to suit a wide variety of contexts. So they can be picked up by parents, by club leaders, by aunts, like myself, or by teachers

1:00:02basically. And so the unit has three lessons. And the first one is called your data in AI. The second one is misinformation in the age of AI. And the third is called using generative AI responsibly. Every lesson comes with a discussion point set, so a set of discussion points. And it comes with two unplugged activities. And the whole idea is that teachers and educators and parents, they can mix and match these activities and discussion points. So they may do a lesson with discussion points only, or they can do it discussions and one unplugged activity or a lesson with two unplugged activities. So it really gives the teacher the freedom to do it in whatever format it suits them. But the whole unit was designed to help teachers have a conversation with their students on misinformation, for example, on how on the accuracy or the lack thereof of an element

1:01:03outputs. So it helps teachers have conversations with their students on how to use AI tools more responsibly and how to set your own principles for using AI tools. So all of these resources kind of help teachers do these activities with their students to really equip them with the tools to talk about these concepts confidently, but most importantly to empower them with the skills to then use AI tools around them in a more responsible way. So there is, is there any training in the experience AI platform? So if I'm a new teacher and I'm picking up a unit and as you've just said, you've got a huge base of units, you're expanding them. Like what is a training available so I can get my head around where I should start with all this? There is, yeah, there's our UK partner who's called Parent Zone and they offer free training for these resources. And I think teachers find it so useful to actually have the chance to sit down and go through

1:02:03those resources with a trainer so that they can ask all the questions. And also when in the foundations unit, the students are guided through how to make their own AI classification system using AI for kids, which is a brilliant activity because it gives them really good hands on experience to see how bias can affect the output of an AI system. And it's things like that that it will give the teachers a chance to try on their own instead of doing it in front of the students and seeing where they go wrong and then they can learn from that. So yes, so Parent Zone offer very high quality training. And we've got some amazing global partners as well. Our resources are, how many countries are they in now? We're 100, 180 countries I think because we have sort of a trainer to teacher to student model where we sort of train the trainer, the trainer then trains the teacher and the teacher then teaches the students.

1:03:04So we've trained the Parent Zone trainers in our resources and then they will train the teachers. So yeah, so it's a really good model works really well. And we had an impact report recently that said that 90% of the teachers felt like they had really understood the AI concepts. It really increased their confidence in the AI concepts through having that training. And then lastly, if I can add as well, we have a training hub which is our own learning platform where we have lots of courses on AI literacy. And I think one of our newest courses actually talks about how to bring AI literacy to the classroom. So it's geared towards teachers specifically. And then we also, we haven't have it of releasing or doing life webinars as well. So we have several webinars on the AI safety unit. We have some webinars on other resources we have written. And these have all been recorded and they all can be found on YouTube as well. So I would say that we have lots of video format resources as well to explain our, we explain

1:04:08the lessons and the resources and where we also walk you through everything that we've created in those resources. And also talking about some of the other things because you had one with Sam Ellingworth and Slow AI, didn't you, Rayhanna, which was really interesting because he talked about how he, how he teaches AI literacy with his students. And one example was that because he realizes that students actually need to be taught how their works being marked effectively so they can create a really good piece of work, he asked the students to ask an AI chatbot to output an essay based on a rubric that he gave them. They then all compared the essays and came to the realization that they were very average and they were all very similar. And so it comes back to what Rayhanna was saying before about the output from AI chatbots can be very, very average in these sort of plasso whereas if they'd done it themselves, they would have put their own insights and thoughts into the essays and they would have been much better quality.

1:05:09So that was a really, really good webinar, really worth a watch. Yeah, so that is the needs of training out there. Yes, yeah, so we have a magazine. It's called Hello World Magazine where we have lots of articles that magazine articles that we write. We have a blog on the Raspberry Pi Foundation website where we release blogs as well on our resources. There is a Hello World podcast where we also interview guests who we think are very, very interesting in the AI literacy space. And then finally, we also have, what else do we have? We have quick reads. So quick reads are, I've mentioned them before as well. These are condensed one or two page articles where we have basically summarized all the research on search topics like feedback literacy. We have a whole quick read on that. We have a quick read on how to avoid anthropomorphism. We have a quick read on several AI literacy concepts including the data driven and rule-based

1:06:10systems that I mentioned earlier. And so we have quick reads on those that you can find on our website as well, which I think are amazing resource because they are short and compact. So if you're a teacher with very little time, you can just have 10 minutes to read them or to listen to our podcasts. And just to butt in, we've also got the big book of pedagogy as well, which is another fantastic resource. It's got loads of stuff about how to teach computer science, pedagogy, go find out, but also about AI literacy as well. I just wanted to ask, it's amazing. I'm just going to say, wow, that's a lot. So good. I genuinely think it's a game change, particularly for people like me. I mean, I never knew this level of content existed for me to be able to get and access with students. So as a history teacher, I mean, and as somebody who may be coming in to, because I think a lot of schools will be looking at this in the near future in terms of how it is embedded

1:07:14across the curriculum, how they are thinking about how they are delivering it in different lessons and also even just a standalone lessons, teachers may be delivering it across the school. So that's brilliant. You know, it's absolutely fantastic. One thing I wanted to sort of, and I'm stealing Matt's big, big finish now, which I feel really bad about. I was so keen to ask it. I'm just going to steal it. But if there was one thing that you wanted every teacher and school leader to understand about AI literacy, what would it be? I think for me, it's that it's so important to empower students to understand it because there's so much hype out there, which makes it sort of feel almost inevitable and it's all very depressing as well at the moment, sort of all the news about AI taking over and stuff. But I think it's really important if students understand how these systems work, they realize

1:08:15that that's actually not very likely and they become much more educated in how what they actually are. And to empower them to ask the right questions, you know, like they're going to go into workplaces in a few years' time and they need to be able to actually maybe push back on some of the AI systems that they're being asked to use or be able to actually use them effectively and then be able to evaluate the output from them. And so really being able to empower students to ask the right questions, to be able to shape the world that they are going up into, I think is really, really important. Yeah, and I would say I'm kind of going to copy Emma's answer here, but I think it's very important that we develop young people into very critical thinkers, but not just being critical of what's happening around them, but giving them the tools to take action as well. Because I think it's very easy for us to, or, you know, I think for a young person, it's very easy to feel discouraged maybe or hopeless. And I think our job is to inspire young people to have the knowledge and the skills to

1:09:19question the systems around them, to change the systems around them. Because at some point, like Emma said, they are going to enter the workforce. And I think it's very, the first, one of the first things that I think is so important to do well is understanding that a lot of these systems and a lot of the systems around us are designed to turn us into passive users. And to be a passive user means that, you know, like it's, it's, it's in the benefit of these systems, if there are more people using them and using them to offload all their critical thinking to these tools, it's all, you know, you're the product essentially because you're using, you're using the tools and you're creating more data for these tools to be trained on, et cetera. But to understand that that is basically what these tools are designed to do is to make you into passive users. And once you reclaim that power back, you reclaim the agency, then I think you can, you can make change. And I think that's very important in our work as educators is to inspire young people to

1:10:20give them the skills to give them the type of AI literacy to then so that they can go and make their own informed decisions, but most importantly to make a difference. And I really hope that with the experience AI resources we can, we can make a difference. It's amazing. Paddy's commented, I'm definitely trying out the flood forecasting unit as it's a great way to help children and staff gain an understanding of AI. Brilliant stuff. All right. I wait. One. I. I. I. I. I. I. I. I thought you looked really enthusiastic about my. Nice. It's absolutely brilliant. I mean, anybody who is engaged with with AI and AI literacy needs to check this out. So come and towards the end, guys, all right. It's been an amazing conversation. I just want to give a massive shout out to Raspberry Pi Foundation, of course, but our guests, Ray Hanna and Emma, it's been absolutely amazing chatting to you this evening about this. I'm also going to give a massive shout out to Matt Wiem.

1:11:23So is over here. There he is. Give him a follow. You already follow. No point. But no, it's been been a brilliant, brilliant chat. Just a final little shout. If you want to check out what we've been talking about here, then you can visit the website, rpf.io, forward slash experience AI, dash teacher talks, and you can find out all about experience AI. I'm sure if you just Google experience AI Raspberry Pi Foundation, I'm sure you can find it there too. And you know, have a look because there's so much in there. And I have to say after talking to you guys tonight, I am really impressed with a lot of the thinking that's sitting behind a lot of what you're doing. I don't know if you are Matt but I am. Yeah. I'm glad. And I've heard of the resources, but I don't think I have explored them as much as I should have done based on the conversations that we've had. And I'm very much looking forward to the new units you've been.

1:12:23Teased to tantalizing as with tonight. I think they're going to be fantastic, especially around that critical thinking piece, which I think is a big word that people are using at the moment. And I like, I think like AI, critical thinking is sometimes getting a bit lost in translation. So I'm very much looking forward to your, your pairing of those two things together. Yeah. If I can add to that as well, which is all our resources are research informed. And that means that a lot of our resources, they use the most backed principles you can, you can imagine. We use a lot of, I think I hope you understand by now that we don't really jump on the hype train, for example, but that we create, we were very, very, very meticulous and very careful in how we create resources and how we teach AI literacy to educators all over the world. So it's not just educators in the UK, but also in more resource constrained areas in the world as well. And I think that's so important about the resources.

1:13:24If we can shout out our own horn, which is everything has been a lot of our principles and a lot of the concepts that we teach have been informed by research and we develop them in direct collaboration with industry experts. So they're all, or they're all factual and they're all accurate as well. Yeah. That's why I can see it. It's great for schools as well. Tech schools and talking about tech and schools. A lot of the resources look like they're unplugged. Like you don't need to be sat in an IT lab or have a tablet or a phone or something in front of you to make them work, which is great for a lot of schools. So important because a lot of the schools that teaching them in like rural Ken, you're in stuff, they don't have an internet connection, they don't have a printer. You know, they might have one mobile phone screen that people are crowding around. So we have to very much keep that in the in the front of our minds when we're making these resources. There's always an alternative option. Say that you don't have to print. I mean, I know that schools I've worked out, the printer has not worked. So, you know, it's so it's really important to have those alternative methods of delivering the content.

1:14:28And I think unplugged resources, especially is so powerful because they really get those sort of concepts across without needing to actually use the technology itself. So yeah. Final, final wrap up, Emma and Rayhanna, is there anything you really are desperately to say about AI and AI literacy in general that we haven't mentioned in this show today? Is there a sound bite that you want to leave us with for teachers, particularly? Don't be scared of it. I think would be my piece of advice. You know, it sounds terrifying. And I know that teachers are again, from personal experience, incredibly time constrained. And you've got your own subjects to teach. And, you know, I know there's a huge amount of work that teachers do. But I think it's so important that we all work together to make sure that our children are using AI safely and responsibly. And they understand what these tools are. And as I said, you know, part of education is making sure that our students understand the world around them.

1:15:29And the world around them is effectively being shaped right now with AI. And so we need to work together to make sure that all of our, all of our youngsters have got that foundation knowledge. So don't be scared. And our resources are there for you, you know, it's all there. You don't have to go and find anything out yourself. It's all been checked by researchers, by researchers at deep mind. So it's accurate. It's up to date and it's all there for you. And it's free as well. Yeah, it's free. Yeah. I would just echo what Emma said. I think she's just sealing my words today. No, Emma said everything I wanted to say. I would just say go check out our resources. We are there to, we are, we are all teachers and we are all educators. We have been in the classroom. And I think we all understand what the, what the struggles are of classroom teachers. Go check out our resources. They're designed for you by you, essentially. And they are free. So we don't get any benefit from, from anyone using them, except that we can, that we feel

1:16:34very humbled that a lot of people are using our resources. So please go use them. Tell us what you think. Tell us what is missing and what is lacking. What you want to see more resources on, because we listen to teachers. We have, we have advisory groups that we regularly get in touch with to get their feedback on pilots on resources that we've just developed or when we want a sounding board to new ideas that we have. So please do get in touch, get involved. We need more people to tell them what is missing, to tell us what is missing. Yeah. Amazing. That's perfect, perfect point to finish on. Thank you, everybody. Thank you for sticking with us as well. Everybody on over on YouTube and LinkedIn and Facebook. And yeah, if you're listening back to this as a podcast, also to you as well. So thank you very much and take care of everybody. And we'll be back on TTR in about 40 minutes. So thanks everybody and take care and good evening to you. You've been listening to Teachers Talk Radio.

1:17:36Tune in live and listen back at ttradio.org. We look forward to hearing from you next time on Teachers Talk Radio.