Meta’s AI sunglasses and that Instagram AI feature

Hello and welcome to Raising Generation AI. Don’t worry, Holly will be joining me shortly. I’m just popping up to let you know that at the beginning of the show, we talk about Meta’s new AI feature on Instagram that they’ve now decided to take down.

So we recorded this on Wednesday before that news broke that they’d rescinded access.

So I just wanted to let you know everything on our social media is all up to date on the latest in AI news. That’s at Raising Generation AI on Instagram, Facebook and on TikTok.

Do come and join us there for the latest. Otherwise, enjoy the show.

Hello, and welcome to Raising Generation AI with me, mum of two, Holly Green.

And me, AI expert, Fiona Morgan.

This podcast is all about the complicated world of AI and parenting because I’m a mum, and I’m a bit worried about it all.

And today we have some sad news. It’s the final episode.

No, but it’s only the final episode of this term. We will be back in September because I still have an awful lot of things to worry about, Fiona. You didn’t think I was gonna stop that quickly, did you?

I have many, many things I want to continue to offload on you.

Funny you should say that. I don’t know whether or not you’ve seen the recent news.

And we’re recording this, we don’t normally specify.

We’re recording this on Wednesday. And the news has broken overnight on Tuesday, that Meta have a new feature. So Meta are the people that make Instagram, Facebook and Messenger and things like that.

You might be familiar with them. And I don’t know whether you’ve seen this, Holly, but I have posted a video to our social about it.

I’ve got this horrible ominous feeling about what you’re about to say, but go on.

Well, you know how we talk a lot about feeding our data into models, and that should be a choice. And you can choose whether or not to share your information or not with models. Well, Meta’s new feature is this AI image generator.

So nothing particularly new or fancy there, but it’s called MuseImage and it’s being embedded into Instagram and Facebook Messenger and WhatsApp. It’s live already in the States, but obviously I’m not in the States, so I haven’t been able to test this out. But what it’s going to allow you to do is to go into, let’s say, your Instagram stories.

So if I went to my Instagram stories, got a picture of me, said, make me look like I’m in a punk band, right? And so it makes me look like I’m in a punk band. What I can then do, and this is the new bit, is I can then do at Holly underscore Green, and say, add Holly to my punk band.

And what that will do is it will scan your Instagram pictures and feed them into the AI image generator to generate an image in your likeness for my story.

Now, go on, not happy about this already, go on.

It’s super new. Like they’ve just put out a press release about it. And I was like, Oh, I need to talk about this on the podcast because this is exactly the sort of thing we should be talking about, right?

But there’s a lot of unknowns at the moment.

Okay, what don’t we know yet?

What we don’t know yet is it’s available in the US.

Has it just become available in the US or has it been available in the US for a little while?

Okay, so just become available in the US.

You’ve been able to remix your own pictures using an AI image generator within Instagram.

So you’ve been able to put in your picture and say, give me a silly hat or whatever. Like you’ve been able to do that for a little while, but what they’ve done specifically is this ability to at mention another account.

And it will pull the images from that Instagram account and put them into the image generator. Now the key bits here are, although it’s only available in the US, I’m not sure whether or not a US account could tag me as a non-US account already, or whether or not you can only tag other US accounts. I haven’t been able to get to the bottom of that yet, and I will update our social media.

So it might even be that by the time this goes out, that there is an update on this. So the social media is the place where you’ll find that out.

Did you say we know when it’s happening in the UK?

No, but they have said they are rolling out to other countries. So I suspect the UK won’t be too far behind, right? We normally get these features pretty soon after.

The other things to be aware of is it only works for public Instagram accounts, so for those of you who have got locked down Instagram accounts, it can’t do it. The main thing to be aware of is that it’s opt out, not opt in. So by default, your settings allow your profile to be used for this purpose.

And if someone does use your account for this purpose, you won’t have any form of notification to tell you they have done so.

Oh, so someone can be creating pictures of you doing goodness knows what, and you won’t even know about it.

Yes. That feels, oh, I agree. I mean, I don’t think it’s something to be hugely alarmed about, because someone could go and get pictures of you anyway, and chuck them into ChatGPT.

So they could be doing this stuff anyway. But what Meta has done is made this so accessible and made it this feature and is selling it as this exciting new feature and encouraging people to use it, right? And brands are going to be able to use it as well.

There’s a whole line saying that advertisers and agencies will be able to tap into this as well. So I don’t quite know what that means. They’re merely looking at their press release and stuff at the moment for the details.

That seems weird because that would imply you could use someone for your advertising campaign or whatever without their permission. So you could imply that someone endorses something when they don’t endorse. That seems really dodgy.

I don’t know if that’s how it’s going to pan out and how they’re going to be able to use it. But the line specifically says that advertising agency will be able to use this.

And I wonder what restrictions there will be in terms of what you can make these people do. I mean, obviously there was all that on X wasn’t there? Do you remember when, I don’t know if it’s still possible actually, where you could make users do things that were not very appropriate?

Where’s the line going to be drawn here? Are you going to be able to make children do things you wouldn’t want children to be doing? That feels very uncomfortable to me.

I think that’s going to be what we see a lot of the headlines and stories be about over the coming days. I think they will have strong guardrails. I would suspect a company the size of Meta are not putting this out without the guardrails.

But I’ve said this many times on here, guardrails are not 100 percent effective. People that want to get around guardrails can usually figure out ways to get around guardrails. There’s also the question of not those extreme cases, but the element of bullying or something.

It sounds like it could be minor, but producer James is a big Arsenal fan. I could effectively bring him into my story and put a Tottenham Hotspur shirt on, which would be an unthinkable thing to do to an Arsenal fan. And whilst it would be hilarious to do that to producer James, I can see a world where it’s encouraging using other people’s images without their likeness and there can be bullying there and it’s fun if they’re laughing with you, but it’s not fun as soon as you’re laughing at them.

So for kids as well and for young people, I think bullying can be so rife. It’s such a difficult time of life, such a difficult experience to go through, I think for many people, adolescents. And I just feel like these layers of social media just give bullies so many more tools, don’t they?

And the way I sort of feel about this, and it’s the way I feel about a lot of technologies is that why do we need them? What benefits are they really bringing? It just feels like a sort of a bit of an invasion of privacy.

And I also feel this way about smart glasses. So, well, you’ll know much more about this than me Fiona, Kylie Jenner has started advertising Meta glasses. And I sort of thought, do you remember Google Glass came and then died away?

I thought at that time we’d sort of as a society decided that we didn’t really like the idea of smart glasses and they would go away and that would be it. But no, they’ve apparently now become cool. And so more and more people are, it seems buying smart glasses.

And I am flabbergasted by this because I mean, I work in an industry where we have to be really careful when we film people, especially children. We need permission from a guardian to film a child. Well, with these smart glasses, presumably people can just film anyone they want.

I mean, I know they’ve got a little light on them supposedly, but I think it’s quite subtle when they’re filming. But I can’t see what stops people going into courtrooms and filming or going into bathrooms and filming. And that just feels like such a tricky issue.

I don’t know how we’ll overcome that. Are we going to see laws coming in to stop some of this stuff? And I’ve also seen that potentially you could have facial recognition in them.

And so you could walk down the street and your smart glasses could be telling you the name of people you’re walking by and other information about them, which just feels, I don’t know, just feels really uncomfortable to me.

I’m sure you saw that I did a little mini explainer on the website about it because I saw that lots of people were asking about why is there so much uproar? And it is exactly that. It’s because with a phone, you can typically tell if someone’s filming.

And whilst that’s still an invasion of privacy, you can still see that. And actually in some countries, taking a picture on the phone, you know that little shutter sound, that ka-ching, ka-ching. You know, when you take the picture by law, that has to be switched on for, so you can’t mute that in some countries for privacy reasons.

So I wonder if the same countries, I’ll have to have a look at it, if they’re having same issues with glasses. So yeah, it’s a really mad topic. Have you ever seen the episode of Black Mirror?

No, I avoid Black Mirror, because I don’t think it would be good for me.

I think you’re absolutely right, don’t. But there is a Black Mirror episode of exactly this.

I don’t think they’ve got glasses.

I think it’s just like an embedded chip or whatever, because it’s Black Mirror. But this whole thing is, you know how on like Uber and stuff like that, you get ratings. This whole episode of Black Mirror is based around humans being rated.

When you look at someone, it automatically tells you their ratings about who they are, whether they’re kind and stuff, and plays out this whole story of what happens when someone’s rating drops. And as you can imagine, it’s pretty dark, it’s Black Mirror, and the impact that has on their life and stuff. And actually, I think we’re watching that real time.

I can see a world where that happens. Even though we’re going to wrap the podcast for a few weeks, we do it alongside term time, and it’s some holidays coming up, whoo!

We are going to carry on doing all of the stuff we’ve been doing on social media, and that includes the AI explainers, like the Meta glasses, the Kylie Jenner ones.

So if you have questions that you have come up, do send them in. Otherwise, I’m selecting them from what me and Holly see in the news. But we’re going to keep doing everything, really, other than the podcast throughout the summer.

We’ve got some AI activities coming up. Now, we’ve been doing these for a little while. Holly has had a go at some of them.

Some are more successful than others. But we’re going to carry them on.

Do you want to hear about my assessment so far about AI activities? Some have been amazing. I loved Chef GPT, where it told me what to make out of the contents of my fridge.

Even though it did try and send me down to the… Yeah, that was it. My sorry-looking bean sprouts.

It did keep trying to send me to the shop to buy additional ingredients, which felt a bit pointless. But anyway, I really enjoyed it in principle. I love the colouring in sheets as well.

The colouring in sheets are amazing. It was excellent. The homework helper was brilliant.
I was really, really, really impressed at how good AI was at explaining homework. So that was five stars for that one. The ones I don’t think AI is quite as good at is the more creative ones.

So we’ve had…

Who done it?

We had the… Yes, we had the who done it, which was where you have… I’m gonna call it a murder mystery.

Clearly, it wasn’t about murder. It was like a who stole the biscuit mystery. Then it gave you clues and you had to identify who in your family had stolen the last biscuit or whatever.

And then the more recent one was the quest one. So you take chores that your kids don’t want to do, you put it into Chatjibiti and it turns into a quest for your kids. And the ideas were amazing.

And I love that you come up with them. I felt like Chatjibiti lacked a little bit of creativity in them. So they were fine.

But I sort of read them and went, okay, I see the structure here. I see what you’ve done here. I think I can do better.

So they were great in the sense that they inspired me to do more interesting things with the kids.

But I was a bit like, yeah, okay, yeah, not a bad effort from Chatjibiti. But I still think maybe the human brain is slightly in the lead when it comes to creativity, especially when it’s something about your own kids, because you know your own kids, hopefully much better than AI does.

And therefore, I think as a parent, you can kind of tailor that better to your own kids.

But it did give me some great ideas. So thank you, Fiona.

I did really appreciate them.

Look, they are a bit of fun, a bit of a way to try and learn a bit about playing with AI and what they can do. And actually, I think that’s a better outcome, really. If instead of just pasting the prompt in, you’re then being inspired to do some fun things with your kids away from the screens, even better.

So they’ll be carrying on throughout the summer. We have six for the six weeks, one every week, and they’re to do specific screen free activities to keep your kids happy and entertained.

If you work for various ages, you can change the age as you’re putting the prompt in.
So give them a go. And actually, we also have a newsletter, which is a weekly newsletter. And if you sign up to the newsletter, you can get all six of those AI activities in one go.

So you can have them for the beginning of the summer, rather than being restricted to the order in which I think they should be released in. So do sign up to the newsletter. The newsletter is just some news updates, anything interesting going on in AI, the world of AI.

So the meta stuff will be the newsletter that has just gone out on Friday, and will of course keep you up to date with all things on how to change your settings so that you can opt out of your data being used for that sort of thing. And we’ll share lots of fun things about AI.

Now you probably remember Fiona as well, that when I had a go at Chef GPT, which I loved, it got a little bit sassy with me, as we know, I’m sorry, licking bean sprouts, giant w of yoghurt, et cetera. And I was telling this to a friend and she told me about an incident where she’d asked Chef GPT to help her write a letter to her kid’s school because her child had been ill and the teacher hadn’t let her know and she was a bit frustrated about it. And instead of coming across in a frustrated way, she thought, oh, I’ll get Chef GPT to write me something nice that I can send to the kid’s school.

Anyway, this is what it came back with. I won’t read you the whole letter. I’ll give you a few lines and I will change the name for privacy purposes.

But this is how it started. Dear Mrs. Smith, I hope this email finds you thoroughly enjoying your peaceful days, free from the burden of notifying parents about their children’s health.

And it goes on and it goes on and it finishes, wishing you continued success in your quest to keep parents in the dark, literally and metaphorically.

Mr. and Mrs. Blocks.

And I just love this. And she tells me she did not ask for anything sarcastic. She was genuinely trying to get a helpful letter from ChatGPT.

So I don’t know what’s going on with ChatGPT, but it definitely has a little bit of a sassy side to it.

That’s amazing.

I don’t think she sent it, in fairness. I think she decided to rewrite it. Probably a good move.

I have a couple of questions for you, Fiona.

Yes.

Firstly, I had a bit of an incident this week with a cake. AI did me dirty because I… Oh, okay.

I was going to say, I’m no chef, no baker.

No baker. I can’t tell you.

If it’s about a cake, I’m out.

It’s more maths related. It’s right up your street. You’ll be fine.

I wanted to make a cake, okay. And I had a recipe, but my tin was bigger than the recipe required. So I thought, aha, good opportunity to use a bit of AI.

So I put the recipe into chat.gpt and I said, can you make this suitable for like a 23-inch tin? Sorry, no, that would be a huge cake, a 23-centimeter tin as opposed to 20-centimeter tin. And it was like, yeah, of course I can.

And it said, you just need to multiply each ingredient by this amount. It did it all for me. I went, great.

I followed the recipe. It had halved the eggs. Everything else, it got the right amount.

Eggs randomly halved them.

And did you blindly follow it and make a terrible cake?

No, luckily, although it would have made a better story if I had have done. Luckily, I did it all and I got to the eggs and I went, that doesn’t seem right. And I went back and checked myself.

And I was like, oh no. So it told me like three eggs and I needed the six eggs. So luckily, no, the cake was ruined.

But I was a bit surprised because I’ve heard stories of AI solving maths problems that humans haven’t been able to solve for decades. And I sort of know what you’re going to say, which is Chachipati is a language model. It’s not a maths model.

Am I that predictable?

But surely that was fairly simple maths. All it had to do was a fairly simple multiplication calculation on each of the ingredients. But no, it just randomly got one wrong.

And it did make me wonder as well. So we’ve talked a lot about ChatGPT, Claude, Gemini. These are obviously language models, but other models we ought to be using and looking at for other things.

And if so, what are they? I feel like this is a whole topic maybe we could go down.

Yes, you’re right. It’s a language model. And these are all language models.

And the way you interact with them is language. But sort of sitting underneath the layer of the language model is other tools, they call them, or access to other models. And so there might be a model that’s particularly good at, I don’t know, forecasting finances, let’s just say.

But the language model does something called reasoning to figure out when to use that financial forecasting model. So let’s say it’s got, let’s actually, let’s use the cake example. So you’ve got the large language model and it has these tools underneath.

And one of the tools might be a calculator, right? A simple, basic calculator. And so instead of trying to do the maths using predictive language, which doesn’t make any sense, it identifies, actually the language model says, do you know what?

I need to use a calculator here. So it calls that tool and uses the tool. It formulates the input for that tool.

So it decides what it’s going to put in. So there’s a step there that can go wrong if it hasn’t quite formulated that. The calculator probably is pretty robust.

It’s just a calculator, right? It’s unlikely that anything’s going to go wrong in the calculation if it’s using a bog standard calculator. But then it takes the output of that calculator.

And let’s say it had six items. I literally don’t bake cakes. I have no idea.

But let’s say there were six items. So it ran six calculations and got these six numbers.

It then has to take those six numbers and formulate a response.

So that’s another step where it can then hallucinate. So what it’s probably done is hallucinated either in terms of understanding what you’re saying, reasoning and deciding what to do, putting the input into the calculator or formulating a response from the output of the calculator. So there’s four steps where that language model is in charge and it can hallucinate.

The problem is that you have access to a calculator. So yes, Holly, someone with your awareness of numbers could have very much just used a calculator.

I mean, sure. I was trying to stop the toddler throwing eggs at the fridge or smearing butter all over the TV or something at the time. It’s understandable.

But when it comes to, let’s say, a financial forecasting model or something like that, which might be another AI model that’s sort of sitting behind one of these language models, you don’t typically have access to them unless you can code. That’s exactly what we’ve talked about before, where these language models have suddenly given access to all this other AI to so many people, which is what’s made it so big. But yeah, you kind of have to just talk to these models through the language model and therefore you risk hallucination.

They are getting a bit better, but they will always hallucinate. They are, as we’ve talked about, just language prediction patterns, and that is functionally what they do, so they will always make stuff up.

So as an everyday person, is our access to AI pretty much limited to the large language models? Is that pretty much what we’re limited to? Because I get the sense there’s all sorts of different types of AI out there, doing all sorts of amazing things, but that’s where my sort of AI knowledge then, well, my limited AI knowledge drops off a cliff.

Sort of. Through other tools as well, you access AI. So we’ve talked about facial recognition today.

So when you go onto your phone and go, I want to see all the pictures of Holly, that’s AI. So you can access AI as well through other interfaces, but someone has to build it. Someone has to make that possible.

And most people at the moment, and don’t ask me my opinion on this, because we’ll be here for another hour, but most people at the moment are just creating new products, which are just ChatGPT with stuff underneath. Like they’re not bothering to build proper interfaces and stuff like that to access it more accurately. They’re just going, yeah, you can just use a language model.

Then everyone can access it and it’s easy. But I do think things will change as we move forward. So you will have access to these other models that do bespoke specific things, but someone has to build you access unless you can code.

I could really do with, I know we talked about this in a previous episode, I could really do with a sort of a PA type organizer to organize my life. So if you could get working on that for me, Fiona, that would be…

Yeah, there’s a few privacy risks there, but yes, sure.

I have a follow up question as well, which has not come in from me. This has come in from three different people who have watched the podcast. Watched, I say specifically, because their question is, Fiona, is your background AI?

And for anyone who is listening or not watching, Fiona’s background looks beautiful. She has this huge window behind her with the skyline of Bangkok and she has this incredibly healthy looking green plant sprouting on her other side.

So, the plant is plastic.

No!

I thought it was real. It came with a flat, so I didn’t buy a plastic flat, but it is a good job it is plastic, because let’s be honest, I have never kept a plant alive in my life. So, it wouldn’t be looking this fresh if it was real, but the background is entirely real.

You might notice sometimes the different weather, but also there’s a building being built somewhere over there, and we see cranes and all sorts of stuff. So, yeah, no, completely real background.

I knew that obviously, but I just thought, I thought I’d let you answer that one for everyone. Now, the other thing I did this week, I haven’t just been baking cakes. I also went to watch the new Toy Story.

Have you seen it?

No, I have not. Do I need to go and see it? Is it great?

So, I love Toy Story. It’s really nostalgic for me. Obviously, we were quite young when it first came out, and I loved the films growing up.

So, I was quite excited to see this one. I don’t know. It wasn’t my favorite.

So, the premise for anyone that doesn’t know is that it’s about tech. So, the toys belong to the little girl called Bonnie, and she loves playing with them, but she gets quite left out and finds it hard to make friends because all the other kids are busy on their tablets. And her parents are worried about her, so they decide to buy her a tablet in order that she can be part of things.

They buy her this tablet called LilyPad. It’s really creepy. And it’s the way then she gets completely sucked into LilyPad and this horrible kind of creepy thing gets all the other toys locked in the garage.

And to begin with, I was like, yeah, I’m on board. This is a good message. However, I felt like, so I’m going to give some spoilers here.

So, you know, if you haven’t seen it, turn off for a couple of minutes. The ending is basically that Lily Pad sort of sees the error of her ways and she becomes a self-sacrificing hero who helps to save the day. And I just felt like it was a bit of a cop out.

So the message was sort of, yeah, tax good in some ways. You just need to sort of use it in moderation. Yeah, sure.

But also this does rely on an iPad suddenly or, you know, a tablet suddenly becoming, suddenly having your child’s best interests at heart and not wanting to sort of suck them into this online world where they can encounter bullying and it’s addictive and yada, yada, yada. And it also required the little girl to realize that real life connections were much more important than online connections, which is a valid point. But I also felt that it was just a little bit of a cop out compared to where potentially they could have gone.

I mean, maybe the problem is there are no answers here really are there. So maybe there was no magic solution they could have come up with, but I thought this idea that somehow this tech sort of became self-aware and saw the error of its ways was a little bit weak.

Do you think they were maybe pointing towards tech companies to be more ethical in their design with it?

Do you know, that’s a great take and if that was their plan and their message, then that’s amazing. Yes. I mean, it may be that…

Well, it may be that that was what they were aiming for. I can’t imagine it’s going to have that much influence on tech companies. But if that’s what they were aiming for, then yeah, completely on board with that message.

Maybe I’m being a bit harsh on it, but I just sort of came away thinking, yeah, sure, yeah, tech maybe is okay. As you say, if we do go down that route of things, becoming less addictive and with more protections for children in place, etc. So I don’t know.

I think you should go and watch it though.

Did the kids enjoy it?

They did enjoy it. I do think there was a slight irony that we’re all watching a screen about watching screens effectively and no doubt this film will get watched and watched over on iPads many, many times throughout the course of the summer. But no, the kids really enjoyed it.

It was, you know, they’re nice, the Toy Story films aren’t they? And I like the fact that Jessie was played the lead character in this one and I thought she was great.

She’s, she’s, you know, girl power.

Yeah, exactly.

She’s, she’s good about Jessie.

So yeah, it was, it’s worth a watch, but I don’t know, the ending left me slightly sort of cool.

But the kids didn’t come home immediately demanding a tablet.

No, and I did sort of ask some of their thoughts afterwards about getting tablets. And obviously my three year old’s a little bit young to kind of have that conversation. But my six year old was, she’d obviously taken on board some of the elements of the film.

She’s sort of seen tablets, but she’s not very familiar with how they work. So I think she’d certainly taken on board some of the elements of potentially tricky online relationships and the engrossing nature of tablets as well. I think she maybe took on board, which

I don’t know might be quite useful in the future.

We’ll see how that all pans out. I feel like we ought to have a quick whiz through the topics we’ve covered this half term. We’ve had a good sort of hop around some different AI topics this term, haven’t we?

Do you have any half term highlights?

I was quite shocked of the AI toys for preschoolers, if I’m honest. I wasn’t really aware that they existed in that form and I couldn’t believe really that they were allowed to exist in that form. I know these tools and AIs everywhere and all of that kind of stuff, but I don’t know, there’s just part of me that just assumed that kids’ toys weren’t there yet, but I was wrong and that was interesting.

It sort of feels like nothing’s protected, nothing’s safe, doesn’t it? And you know, AI can infiltrate anything and it doesn’t seem to be that much to stop it really. If someone’s got an idea and can make some money, they can do it without many regulations in place.

I don’t know how it feels, maybe I got that wrong. No, I agree. I’m definitely not going to be investing in an AI toilet anytime soon.

I really like the What’s AI Doing to the Kids Planet episode. I thought Dr. Kelly was great. I thought she was very knowledgeable and very interesting.

And I think lots of people really care about what AI is doing to the planet and how they’re using it. And I think maybe that’s hopefully giving people some food for thought. I think for next term, I would like to maybe dig into AI glasses a bit more, find out a bit more about them.

And also, I still think we a little bit of a dig into some of the more creative side of things and the effect it’s maybe having on the creative industries, art, writing, or just keep updated. Do you remember I was telling you about my friend’s book club? My friend who is a journalist slash writer is in an AI book club.

So she writes stuff and her friends get AI to write things and they compare and contrast.

They most recently did a travel blog and this time they did it all anonymously. So all of the friends got AI to write something and she wrote something herself.

Then they all had to vote on which was the real bit of writing and which was AI. Do you think people knew which bit was the real writing?

I have faith. I think they would.

Yes, thankfully they did. Everyone said the AI writing was much more generic. It didn’t have a voice in the same way like that personal voice that my friend’s writing had.
So yes, thankfully everyone could tell still which was the real writer and which was the AI.

So I thought that was a bit of a relief. But yes, I think looking a bit more at those creative industries, I think would be really interesting next term.

Have you seen the other creative AI news this week as well? Which I think you’ll be absolutely horrified to see.

You saved all of today, haven’t you?

There’s an AI actor, an actress, who is very openly created by AI. It’s not something where they’re trying to trick people or whatever, but she can be cast in movies and things like that, she’s being pitched and put in front of casting directors in the same way humans are and things like that, and she has got her first part in a movie, so there is a production company that have done that.

I can’t even in my mind work out how that works. How do the other actors act with an AI actor? It just makes sense to me.

Surely one person is AI, they all need to be AI, don’t they? In which case, just make the whole thing AI and hopefully it’s-

Maybe it’s a bit like CGI.

Yes, but when it’s CGI, don’t you still sort of have something there, like it’ll be like a person dressed in green or something, that the other actors are acting towards. If it’s just a huge- I don’t know, that sounds very odd to me.

It’s mad. The world is crazy. We’ll cover all this stuff.

We’re going to keep going. We need to keep going. The world of AI is crazy and we’re going to keep going.

We’re going to take a bit of a summer break from the podcast and then we’re going to come back and there’s going to be a bazillion news stories for us to discuss. The Prime Minister will probably be AI by that point.

You joke, however. No, it does. It’s the speed at which it moves and changes, isn’t it?
I feel like since we started the podcast, things have changed sufficiently that I now have new questions that we need to answer. So I feel like it’s going to be one of those podcasts that almost never concludes because there’s always new things to consider and look at and talk about and think about. So at least that will keep us busy, won’t it?

Now, I think that’s just about it for this episode, isn’t it? We align to the school terms, so we are going to have a bit of a summer break now and come back on the 7th of September. So I think that’s what, seven weeks away, is it?

No doubt all sorts will have happened in the world of AI by then. We shall see. All sorts of horrible news for you to come and greet me with in September, Fiona.

Well, don’t worry, I’ll be keeping it all over the social media as well. So you can follow along there. We might be taking a break in the podcast, but there’ll be lots going on, including the Screen Free Summer Series activities.

So do let us know how you get on with those on social or on email podcast at raisinggenerationai.com.

So have a fabulous summer and we shall see you at the start of the autumn time.