Episode Transcript
[00:00:00] Speaker A: Welcome to Within WordPress, the podcast about WordPress and all things around WordPress. With me today is Alex. Alex, welcome to the show.
[00:00:16] Speaker B: Thanks. Thanks for having me. It's. I've been, I've been watching, listening, whatever you do, but never been a guest on here. So thanks, thanks for having me on.
[00:00:23] Speaker A: Well, this is why we go to events, so I can pressure you into it. And at the event, Alex, for those who don't know who you are, please introduce yourself.
[00:00:34] Speaker B: Hi, I'm Alex Moss. I am from Manchester in the UK. I have been working in SEO and WordPress for a very long time. SEO more so been doing that since about 2000, 1999, something like that. I've been working with WordPress since about 2011 12. So I consider myself a late bloomer even though I've been in it for nearly 15 years. To me, I came in late and did discovered it going, how the hell did I not understand that there was a great CMS out there? So I have my own agency which I opened about a year after that and that still trades. It's just a small1, about 6 of us remote, but everyone's based in Manchester. And as well as that, I'm principal SEO at Yoast, so I dig in between product marketing and product innovation. So a lot of people inside Yoast have to deal with a lot of my ideas, some of which are terrible, but some of which I guess are
[00:01:24] Speaker A: okay and pass through, some of which are terrible. I want to hear them. What is, what is the worst idea that you came up with? And they were like, nah, we're not having it. But you pushed.
[00:01:35] Speaker B: I don't know.
Usually it's something about UX where I say, can this be done? And then it turns out that there's a legal or regulatory issue that makes me not want to do that and can stop you from doing that. So it's all of those. I'm not, I'm not as not. I wouldn't say I'm not risk averse. It's just I don't even sometimes when I do product innovation, I don't think about the risk until later. That's a question down the line. You know, risk legal, reg, all of those things. I just think of something, then some poor schmo has to go away and see if it's actually legitimate, that can be done.
So I don't envy that person. But yeah, lots of, lots of ideas. Idea machine gun. Some people call it idea diarrhea. You know, it's. It comes out, but it has to I also. I call myself 10% because only 1 out of 10 of my ideas are any good, but I need. I need to get the 9 out in order to get the 10th idea.
[00:02:28] Speaker A: I recognize that. I recognize that. I think it's also kind of a muscle. The. The more you let them out and discuss them, the better refined they are, whether that means it's a go or no go. But also, hey, that one gave me the idea for the next one.
I think I recognize that.
So.
[00:02:49] Speaker B: Exactly.
[00:02:51] Speaker A: So Yoast. You work at Yoast. I worked at Yoast as well.
So I kind of.
[00:02:57] Speaker B: Not together, though. Not together.
[00:02:58] Speaker A: Not together. No, no, no, no. We didn't think that was safe. So most people know Yoast as the Yoast SEO plugin.
You described your job as Principal SEO. What does that mean?
[00:03:15] Speaker B: And principality. I even had to look it up myself. It's more of a. Before IC was a term, it was kind of an IC thing. You're not necessarily a people manager, but you know the technical ins and outs of that particular skill set. So for me, I am one of. Well, there's Carolyn as well. She's my counterpart in the US and we both work together to make sure that everything's best practice, everything's kosher, anything that's getting deprecated or not looked at anymore on the side of search, we come in and we make sure that everything's up to date.
[00:03:45] Speaker A: So.
[00:03:45] Speaker B: And sound and built well. And as well as that, we think of new things based on what's around us and what's going on, as there is a lot going on around us. And it's actually quite hard to decide what to focus on, what not to focus on.
I mean, again, you'll probably relate. I've wasted a week doing something which you think is going to get somewhere, and then at the end, you kick yourself for even spending a day on doing that thing. But again, it's like the ideas you have to do it. I think to me, openclaw is the example here. I wanted it to work so, so much, but I was finding that I was spending 10 times the amount of legwork to get things set up the way that they should just to. I could just do that thing. I mean, it's not. I don't see anyone who's come out and said, I installed OpenClaw, and three months later, I'm now living a different quality of life. Not one person can claim this yet. And until they do and give me an actual use case that isn't X saying there's Passive income and then the next twee tweets about crypto or something. I just.
I'm skeptical. I'm a real big optimist, which is why I feel like I'm an early adopter of all of these things. I just want to see everything. Can it work? Can it work for me? Can I think like my wife who's on the creative angle of digital marketing, could that thing work for her?
And a lot of these things are. Well, they kind of could, but do you really need them right now? Is that the actual solution right now? You don't need everything right now. Things are. Things are coming. I would say things are coming.
And that's the exciting part. What are those things and how are they going to differ to today?
[00:05:19] Speaker A: There's a lot of excitement if you're daring enough to look at it that way.
I think before we started recording I said one of the things I want to steer away from as much as possible is the doom and gloom mentality. I don't think that benefits anybody. I think it's much more. I think conversation should much more be about opportunities seen and looking at ways to better yourself in your craft and all that sort of stuff. I think that's a much more interesting approach to life in general.
But we do live in interesting times with the advances of AI. I like your example of OpenClaw. Obviously I've installed it in an entirely no access to the Internet type of way, no access to my client stuff, no access to my personal stuff, nothing. Because I just want to understand what it does. And like you, I don't see the advantage. And I've. I've watched quite a few YouTube videos, read articles and I go, I get what it does, it's just not. I don't. I don't see how that's going to materially change my way of working.
[00:06:30] Speaker B: Well, I guess more so, I think it's the theory of it that's going to change. So without OpenClaw, there wouldn't be scheduled tasks or loops or would there? You know, inside Claw, for example. It's just. I think that would have happened anyway, but maybe openclaw accelerated people inside Anthropic's offices to do more quicker. And I think that's this race of getting stuff out fast that this year I think, not haphazard, but there's. I feel like there's a new service, a new app, a new thing to do, something coming out every day I go away, a week on holiday, I come. It's as though I've missed a year. It's. It's a lot. It takes you two days to catch up with the week. And in those two days something else has happened. So you've got to keep on catching up. But again, you've got to pick your battles. But I mean, I agree with you as well. I'm really optimistic about the future, the doom and gloom. I listen to all of it. I want to hear that point of view.
I think the closest one, and I don't know who it was who talked about there would be 15 years of dystopia, but once that 15 years is through, there would be utopia to do with humanity. And I think it is the confusion. And it's probably similar to the way that the Internet came out. People were a. It's a phase, which isn't happening here. We understand that this is not a phase and it's something new. And so there's. Therefore there's not a lot of trust. And we need to get a lot of trust earned by humans in AI, otherwise we're going to keep thinking this is Skynet or the Matrix and that's our thought. We're paranoid that something clever in us is going to overpower us and use us for fuel or work or whatever. And I just don't think that's the way they think. That's the way we think that they think.
[00:08:04] Speaker A: Yeah, I think I agree with you.
It's changing everything, but not necessarily in the way that it intended to change the thing.
And we're learning as we're going.
Just to reference openclaw, again, I'm not actively using it, but when Xai, was it this week or last week? It's one of those things, right? There's constant noise of new things being introduced, but for Grok, there is now Grok bot. And I look at that and I go, now hold on. This is something I feel like it's safer for me to use. It's also more versatile for me to use.
And then I see other people come up with the first ideas of how. Because first, for those of you who don't know what GROK bought, it is essentially an automation, plus planning plus loops, plus, you know, understanding context of where you are, what you need to do.
It can. It can do stuff on websites, interact with the web pages, fill out stuff and all that sort of stuff, but not much from a developer standpoint or a command line type, more like anybody could use it.
[00:09:21] Speaker B: And that's the way to go because we've been building stuff meant for us developers are building for developers and if everyone else, which is open from it, so be it. Yeah, which is open claw. But like you say, openclaw makes other people think differently and create something that's To a different audience. I mean, I know people aren't going to use Grokbot simply because of the connection to Elon Musk and people's political views or thoughts on him personally. And even then I think that's a loss, like maybe, maybe really good opportunity for that person.
[00:09:50] Speaker A: You know, anybody coming up with that moral high ground. I say, is there any social media or big tech company whose leader you fully align with? Because I have a list of a few and I'll, I'll give you stuff they're doing which you're not going to agree with. You don't agree with any, anything of anybody at that level or power. So I, I much look at it from a utility utilitarian. Is that, is that how you say it? Yeah, a point of view, like, what does it allow me to do without compromising me?
I don't know.
Buying into a system I don't fully agree with.
[00:10:30] Speaker B: Exactly, exactly. But people think there's bias in, in LLMs now. Right. So. And there is, you know, without. Yeah, there will be, based on the leanings towards what however it's been built and however it's been trained. So people may think that Grok as an LLM may have, you know, the political and personal opinions the same as Elon Musk, which I don't think is actually true anyway. I think it just gives a bit of sass, doesn't it, compared to all the other LLMs. So. But again, that's part of. I'd rather deal with sass than, hey, that's a great idea. You know, every single time. Not everything I've got is a great idea. I just told you that.
[00:11:07] Speaker A: No, no.
I saw a benchmark not too long ago of Gemini, Claude, ChatGPT and Grok. And Grok was the only one on the political spectrum. He was kind of in the middle.
[00:11:24] Speaker B: Yeah.
[00:11:24] Speaker A: And the other ones were quite on the left side of things, for there was very standard questions to ask and the answers were just essentially giving you the political lay of the land of those tools.
If, if I want to choose a tool to use, I want it to be as neutral as it can be. So yes, that sort of made me go like, okay, I like, I like hearing opposing views, I like understanding opposing views. I don't necessarily immediately agree with whatever that thing spews at me, but I'd like to be offered the option to think about it.
[00:11:58] Speaker B: Do you do that with your LLMs? Now where in my perfect world, I would have kind of. I don't know if it works like open routes, but where you have one central.
You have one central session and it either chooses or it converses with each other with other LLMs. What I do now, sometimes I think of something in Claude. I get a certain way through and then I open up Gemini and I either give it the same question or I tell it to critique what Claude just said. I get the critique, I go back to Claude. So now I've created. And you can't really do that autonomously with things like openclaw yet. But it's like having a team and consensus, isn't it, between LLMs.
So if one LLM may have one opinion, another LLM may have another opinion, they're closed off from each other. Isn't it nice to get the conversation between those two LLMs, not just two agents?
[00:12:45] Speaker A: I do it all the time and I think cursor kind of allows you to do it because there's an auto setting that it can go to whatever LLM makes most sense. I don't think we have the control yet to do exactly this, but you're right, I do that all the time because I want to hear the other side. I want to hear the angles. I don't get the same way. When I write an article, the end result of my draft goes into whichever LLM and I'll ask it, what am I missing? What's the logical fallacy? I'm jumping in, obviously grammar, because English is not my first nor my second language. I need checking there.
And it retains my style, it retains my message, but it sometimes adds two or three paragraphs ago. Oh yeah, crap, that. That should have been there.
Yeah, it's an aid. And then the output of that is then put into the other one because check, double check and check again is how I'm being raised so.
[00:13:43] Speaker B: Exactly. But I don't. Would normal people do that? Or is that just because of who we are and how we work and what we do and the way we think? That's the. That's why we're getting a third. A third perspective, third party perspective from another LLM.
[00:13:56] Speaker A: Yeah, I don't think most people do that.
[00:14:00] Speaker B: No. But it was also interesting because my parents, who are completely technically inept, right. They hate everything to do with technology.
Both my parents now use Gemini daily and there's a couple of reasons I figured out for that. They're on Android. So Google's going to push their own products. Fair enough. They're already familiar with the Google landscape. It definitely gives you a tone of voice that's googly in all the feedback and it does do all the research that my parents just. They now think of the notion of doing your own research, which is what we've been doing all this time. Idiotic.
So you're telling me that if I wanted to buy a car, I would have to spend hours and hours and hours looking at tens of websites going onto this forum, checking. Now it's all done for me. And I don't care about the trust element. I'm sure it's right. And if it's not and I want to double check, that's the point when I'll go and do my own research or I'll go into a car showroom or whatever. So yes, it's changing the way in which people do research in general, but again, there could be fear of the people. Just like a calculator. Is this like a worse version of the calculator where you know the answer, but the important parts, the working out in the middle, what happens to that working out now that these agents are there doing, doing the middle part for us?
[00:15:17] Speaker A: Yeah, there's a risk there. If, and I think this is, this goes back to the same principle.
There's folks who are abandoning WordPress now because AI allows them to build sites.
And there's tons of stuff why, tons of reasons why that's a great idea.
But from my perspective, there's way more reasons why that's not a great idea unless you fully understand what you're working with, as in the output.
And that means like literally everything in the output, not just the technical SEO side of things. So you have valid HTML doing the right things for SEO, all that sort of stuff, but also from accessibility, from link route and all that sort of stuff, all the stuff that you would normally have inside in CMS to guarantee quality of output.
You're also siloing your clients into your output. Right. So there's no way the general client is going to add a little paragraph here, change a little this there. No, they're all dependent on you. And if that's your business model, I mean, more power to you, I guess it's just not my way of working.
I think it's.
[00:16:37] Speaker B: No,
[00:16:40] Speaker A: I think it's interesting to see how people use AI and I think.
So you come from an SEO perspective, meaning you understand Search Engine Optimization, but that acronym still assumes you understand how to search. So take the example of indeed my parents, when they're looking for something, they're just really low tech going for the thing they're searching for. And when I search, I do the exclusions, I want to focus and I filter more and all that sort of stuff. And we see the same behavior with AI, like, if you have a rudimentary level of looking at how do I retrieve the information I'm really looking for versus what the vast majority of people do. That's a big difference. How do you, on the Yoast side of things, how do you keep that in mind? How do you play with that incredibly wide variable?
[00:17:47] Speaker B: Well, it depends which is the famous SEO answer for everything. But it does depend on each business model. I mean, everyone should have their own ideal candidate profile Personas, which we have. We have five different ones, I think maybe six now.
And then we just try and cater as much as possible to all of those ICPs and that's it. So when you were talking and you were saying like, you and your mum, different people, I'm like, all I was thinking of, yeah, but I've got to get both of you on board to get to my site and convert you, both of you. And how do I do that knowing that you're both very different?
So in Yoast, your mum is of no interest to me. You are of interest to me. So I'm going to prioritize content and tone of voice and targeting generally to someone like you. And hopefully you would influence your mum to then look into it and they would get some other perspective and then purchase off the back of you. So the word of mouth is still there and the brand authority, but you're more of an authority than your mum in WordPress. So hearing you say Yoast is the best WordPress plugin out there, then it would have some authority and go further, you know, so just say that and then I'll be able to cap cut it at some point.
[00:19:07] Speaker A: Yeah, well, first of all, I'll have you know my mom is on WordPress, she has a blog.
So you are.
You should also treat her as a customer. But you're right, there is a.
The word of mouth doesn't go away.
We've already talked about AI from a few different perspectives, but I think one of the biggest tensions we currently have is the understanding that working with an AI of any sorts takes away a lot of the actions you would normally perform inside a search engine interface.
First of all, is there any data on what the actual percentage of people who have moved to a LLM for their queries primarily versus what used to be on going to Google or Search Brave or Bing or whatever,
[00:20:12] Speaker B: bits and bobs. But by the time the data's out, there's already a shift happening. This is weird. I think literally since ChatGPT came out, I've been in talks where you can't even do year on year data. Next year I don't think the data and the visitors and the way it's shaped will be the same as this year, which isn't the same as last year. So it's very hard to compare this change. But there are more people, I think that everyone that you know somehow will have used in, you know, some form of LLM once.
I'm talking about my parents using Gemini. So it is changing. And Google did say search is a function.
All I know is there's a few stats that are coming into my head. One of them is that only 16% have opened an LLM ones and the rest of the 84% of the planet have never interacted with an LLM. Right. So we're already in that. Yeah. So it may be 1516 now, but it's not going to be 20 yet. So that's about adoption taking a while. When I was explaining the Internet took 10 years to get used to it, I think the technology is going faster than us and humans are still going to go as slow and we're still going to have to build trust. The other stats, I can't remember them exactly, but Kyle Risley who works at Shopify, he wrote a blog post only in the last week or two and it was all very positive from Shopify's point of view that there were different types of visitors coming to the website now. They were usually more, had more intent and there were higher value, but there was a lot more happening now within LLMs, I think. I don't know why 197% is popping in my head as the increase in LLM consumption.
But the blog post, I'm sure at the bottom right now when you do the replay will be there.
Yeah, I'm seeing just positive stuff and like you say in SEO, it's a bit threatening to some because search as a function is changing the 10 listed links in a SERP. I've not seen one in a while.
I don't know about you, I'm not really. I'm in Google doing some stuff but it's usually not a first interaction with something or it's something super simp, you know, the weather. I'm not asking Claude what the weather is. I'm not wasting my tokens on that.
And that's what Openclaw eventually did for me. I spent 10 hours setting it all up and it gave me a morning digest, including the weather, which by day five something went wrong and it couldn't give it me. And traffic, it's one of those things. But I think those are the positive things that we are changing. We're getting used to it faster than we were with the Internet. It'll force us a lot as devs to work faster to do and adapt to more things. But I think the future of websites, the future we con, the future of how we consume websites, purchase is all going to change just as, just as dramatically as the high street, you know, and the threat to the high street because of online and it took a while and the high street is still there because people still like that thing and it will still be there in 20 years, but it will shift again. Yeah, we're all going to be all about in. I think in real life experiences are going to be the most valuable thing.
There was one more video I watched when someone said if you saw, if you showed a child now a video of Niagara Falls, would that child believe that Niagara Falls is real?
Right. Just because you showed them a video. Right. And they won't know. The only way to confirm it isn't by going on the Internet because it could still not be. It's by seeing it with your own eyes. So things like travel, I think it's going to become hugely important, more important to us as humans, you know, in the future as things that are around us offline are going to maybe I feel like it's, you know, when you come out of a club, you've been in a club all night and then all of a sudden your ears are all ringing. I think it's kind of going to be like that, going more offline because we've been going offline offline. We came online, now the social media, it's at the top. And I'm, I'm hoping, my optimism in here is that the, the state of influencers and you know, TikTok will change for the better and that we may evolve away from doom scrolling. Maybe doom scrolling is going to be a term like cassette.
I'm hoping that's going to be the case.
And this is a weird, murky area that we can't see the light out of the tunnel, but I think there is going to be one. And that right now is maybe the worst of what we're seeing. Which is weird to say because I think people are saying this is the best of what we're see. I seen. We've seen nothing yet.
[00:24:29] Speaker A: Oh yeah, I'm. I'm with you 100%.
I really like the stepping out of the club. First of all, anybody under 30 who doesn't know what that is, you.
[00:24:39] Speaker B: Because that's it.
[00:24:41] Speaker A: That's already a difference.
When you go to a club, there's a lot of loud noise. And because you've been exposed to a lot of loud noise when you step outside after, you know, 4 o', clock, 5 o' clock in the morning, the absence of the noise makes your hair, ears, ring a little bit.
[00:24:59] Speaker B: It does. Maybe. I'll tell you what everyone has done. I'll tell you everyone has done. Woken up in the morning. If ever, if anyone has been woken up by someone just quickly opening the curtains.
[00:25:08] Speaker A: Yeah.
[00:25:09] Speaker B: And then you have to get used to the light. I guess that's the vision version of what we're explaining here. It's stinging at the beginning, but eventually you get used to it and you get your vision or your hearing back.
[00:25:18] Speaker A: Your, your senses adjust is essentially what it is. And there's a delay.
Everybody has that delay.
I've known moments where it's almost. This is the closest I've ever felt to be in a simulation. It's almost as if the screen, my eyes, is building the picture as I am processing that sort of stuff. So I think that's a wonderful analogy with the technology. It's not just AI, but it's a large driver. The technology being injected into our core existence is, is also going to be the one that is making us yearn for anything authentic. Whether that's a conversation, whether that's a, an experience or a travel or, you know, any of those things.
[00:26:13] Speaker B: I mean, or creativity. Anything. That's forward thinking. I've been listening quite a lot to, you know the actor Joseph Gordon Levitt.
[00:26:21] Speaker A: Yeah.
[00:26:21] Speaker B: He was in third from the song Batman. All of that stuff. He's worked with Nolan.
Yeah, yeah, he, he has. Just look at the videos of, of him. He's. He's coming from AI, from the Hollywood perspective and creativity, making movies, understanding what's happening, privacy, humanity. He's really interesting and he kind of sees the sweet and the sour of it, but he sees an opportunity and everything. Like you were saying before, I don't think we should all be scared. We're just thinking, well, how do you solve this thing, this problem that might be there for us, that we're not used to. And he's solving the money royalties in creative. And like you say, we're going to yearn for stories, which is something an LLM cannot do. It synthesizes what it knows already and gives it what it thinks. And we've all seen those videos, those AI videos where people are having a conversation with each other. And it reminds me of Mayor Quimby from the Simpsons. You know, it's all monotonal and everyone speaks the same like that and there's no emotion and everyone goes, oh, my God, you just killed my son. Here's an ice cream. And it's just. There's no change in that. And maybe, maybe it might get. It will get better. It will figure out all of those things. But the true. The simulation that you were thinking about, the real life around us, it's just not. It's not going to get there for a long. I don't think. I think it'll take a good 20 years before we go to a cinema.
And we literally cannot tell the difference of whether that was AI or real or not. I don't think we're there yet. I think we're more than five years away. It's going to look really good. But you and I both know the way that birds fly, the way that water flows, all of those things that are going to be so hard to reproduce and take so much compute power to make it affordable.
It's going to be a hard one. But again, that's where he comes. That's where Joseph Gordon Levit comes in and finds that stuff really interesting. And Ben Affleck, he sold a massive AI company for stupid amounts of money.
[00:28:14] Speaker A: Yeah, yeah. He. Ben. Ben caught a glimpse of what it could be. Built something around it and then sold it for ridiculous money.
[00:28:22] Speaker B: Yeah, it's. He knows what he's doing.
[00:28:24] Speaker A: It's inevitable. It's going to be, I'm tempted to say, sickening. And amount. Sickening amount of determining by AI what our next version of society and websites and interaction with content and all that sort of stuff is going to look like. I say sickening because it's. It's not necessarily in the. In the negative version, but in the. It overwhelms you when you really start to think about.
I mean, we're on this journey for two and a half years on the AI side. Three, maybe three.
We've been building websites for. In my particular case, since 2006, professionally.
And there's a certain forward projection we can do based on the past experience.
And that's the only thing we can actually use now. Those 17 years, plus the three years of AI and you try to turn that as much into what the future is going to look like. I think if you really, really put your mind to it, it is, it is, it will turn your stomach in, in the sense of.
It's just so much, it's just so much that will change.
And, and you know me, I'm, I'm from 73, so that means I live through the 70s, 80s, 90s, which is as analog as it gets. I've seen the rise of technology. I got my first phone 96, and I was an early, early adapter.
[00:30:05] Speaker B: Yeah.
[00:30:08] Speaker A: There's not a kid out there now that doesn't have access to either an iPhone or, or an iPad type of device.
At the age of around 10, they will have it.
That is so fundamentally changing how the next generation is going to look at whatever we're doing that we certainly me, I, I turned 53 last week.
We can't comprehend what the, their minds are going to come up with because it's just too far.
I find that a very interesting thing to look at, but I, to bring it back to what we were talking about, I think we will end up with a yearning for the authenticity the, the connections. Which is why I don't, for instance, I don't think events, whether that be webcams or WordPress related or tech related or whatever related. I don't think they're going anywhere. They might change, become maybe even less popular.
But that's a connection we need.
[00:31:16] Speaker B: That's us. That's what we've always had, is the human's human. If we take that away, then unfortunately it all goes away. But like you're saying how. What is it going to be like for kids when they're older? Well, they're growing up where this is all normal, you know, like how we grew up. Well, you were saying you saw analog. Right. But we adapted to it when we were younger, unlike our parents who. It came.
The Internet came out for our parents, how AI is coming out for us now and they're going to be more used to it and more au fait. And soon, you know, your kids are going to say stuff. You're like, whoa, that's interesting and clever and I would never have thought of that because you're into it more. And maybe, I don't know if you've seen the social experiment on Netflix. It's quite old now, but I remember distinctly. Yeah, so they invented the guy who invented the like button. And he just thought, there were two things that I learned from it. One was they didn't mean to make what eventually happened. The guy just wanted to make a like button. That was it. He didn't realize that what it would do was potentially actually change the way that people had mental health issues. It could cause someone to commit suicide. You know, all of those just wanted to make a like button. That's all it was about. And the social dynamics around that was weird. But also in hindsight, the question at the end, when they were all like, do you give your kids social media? And they're all like, no, no, go nowhere near it, right? And I wonder if our kids are going to understand that at that hindsight when they're older, learn from our mistakes. Because let's face it, our generation and the generation below us in between them, we've been put on a bit of a pedestal. We've been given an easy time. I think especially the Internet's given us easier time. Like you're saying, what's going to change on the Internet? How's content going to be consumed? I'm excited about that because it takes away mediocrity. So when before the Internet, you had to go through journalism to write an article. Now anyone can spin up a website, write something, and now what? You're a journalist for a small publication all of a sudden, day one, it's not going to work like that. In the future, I think you're going to have to earn your authenticity rather than just say, it's there and it's there. Which I think is what we've been all been able to get away with up till now. Maybe some of us work harder than others to get that authority, but there's definitely, you know, there's enough people in your mind probably now, and you're thinking, and the audience is thinking of at least two people, like, how did you get there? And you didn't even work very hard. You know what I mean? And I think that maybe it'll be easier now, but then it will be harder because you're going to like the calculator. You're going to have to know the figuring out in the middle.
[00:33:40] Speaker A: I think that's a good one. I think that's a good way to be looking at the change.
We're essentially in the middle of it, right? So we're already.
When I'm on LinkedIn, it's probably the platform where AI is present the most in terms of the content written there.
I actively disengage with Ads that I see that I really don't care for. So I'll hit that button and I say I hide this ad and it'll stop showing up.
I'm not necessarily against ads, but please have them be relevant at some level.
And in doing so, I saw that there's now a button. Sorry, an option. When you click that button, it says this looks like AI slob.
[00:34:32] Speaker B: Yes, yes, I noticed that. I went on LinkedIn and I don't see it and I haven't checked since. So I do need to check again. But I saw it on X that people were chatting about it. But I like that that you're getting reported.
[00:34:46] Speaker A: I'm like, this is the first sign that people just had enough.
I post something on LinkedIn and I get a few comments and then as soon as there's a few comments in, I get one of these obviously AI generated type of comment, which is half rehashing what I said in my original post and then bringing something quirky or information adjacent, whatever, and I go like, what are you thinking when you're allowing a tool? Because I'm hoping it's a tool and you don't actually manually go in and post it. But what are you thinking that that adds or contributes to whatever I'm saying or somebody else? It's not necessarily about me, but more like how does your mind work where you think that's the authenticity you need to display of yourself? I don't get it.
So yeah, I delete them and. Or use this looks like AI slop. And then if.
And sometimes I'll call them out on them, but never the reply you're hoping for.
[00:35:53] Speaker B: It's interesting because LinkedIn is important some people and they do waste it with AI slot. By the way, I did have a check and I can see the this looks like AI slot button now. But it is interesting because that's like going to an. You know these are all digital entrepreneurs, right? Do you know what? Okay, give me your phone for a day and I'll start calling people as you like, clearly. And I'll be. You don't worry about it. I'll be in your tone of voice. I'll say it the way you want me to say it. Give me your phone now. No, of course they're not going to give it you. I just don't understand. They leave it on autopilot. I think DMS are quite annoying as well. I'm getting them. I'm getting ones like choose 1, 2, 3 or 4 based on what it's like I didn't call a call center. Why am I choosing 1, 2, 3, and 4? Based on what? You, you, you message me like, I don't understand. So I answered 97. And they still, and you could see it was still a bot. Was still trying to figure out, like, I don't quite understand that you're like. Because you're not human, you don't understand humor, sarcasm, you know, the creativity part, it just doesn't. I know, I know what you mean. I, I, I've tested it a few times on LinkedIn. Give me a post. And no, like, I get it. It makes me think of how I don't want to do it and then do it that way, but at least it gives me something to, to start from because as we all know, a blank canvas is the worst thing. I'd rather have a bad canvas than a blank canvas to work from sometimes.
[00:37:16] Speaker A: Fair enough. Stuff.
[00:37:18] Speaker B: You're wrong and I'll tell you why.
[00:37:19] Speaker A: Yeah, well, this is exactly what a friend of mine from the Serbian WordPress community does. Alexander.
He, he engages with all those AI, clearly AI written posts and comments, and he calls them out, every single one of them. I don't know if he still does it, but I've seen him do it for months.
And I like what he does because he's basically saying, look, this is a very cleverly written AI stance or position or comment, but it's clearly written by AI and it misses the entire context of what is inferred here and this and that. So I can probably guess that he uses AI to clarify what is missing here. Maybe not, I don't know, would be hella ironic. But the end result is that he has a, A, a, a, A mode where he looks at it and he goes, okay, this is not, this is not you, this is not real. I'm gonna call you out on it as well.
And I, I like that. I don't want to put the energy in that, but I like the principle of doing that.
And when I look at my kids, their AI detection levels is way higher than, for instance, my parents.
So it, it and, and my kids are 19, 26, 28.
So.
I think there's a sort of maturity to understanding how to use it, but I also think we're just trying to figure it out. Like, there's too many.
[00:39:02] Speaker B: I'd love to know how your kids are using it. Like, my child's only four and a half, so it's not quite there yet. It's not, you know, using an LLM, but I'd love to know how, you know, how are the children using it, you know, beyond studies, because studies is obviously predictable and we would, I assume we all know how to use it that way. But what, what else, what else are they doing that you've thought, oh, maybe I could adopt that tactic or thought process in my, in my work life
[00:39:29] Speaker A: so I can give you some answer on that.
I try to be a parent who gives them a direction of how to look and go about gathering information, then forming your ideas and then challenging those ideas to form your opinion.
And when AI became more prevalent, I assumed they were using it. And I basically had a conversation with all three separate, essentially giving them a. This is the best way to use it because there will be hallucinations, there will be assumptions. There will be. And this is because this is how the world works, not just AI people work with the same types of assumptions which can turn into hallucinations.
And what I see is for the more complex, like how this does this interplay type of questions they have.
Those conversations would very often I would see a part of it because they'd ask me about it.
Now I get this is the result I already got. How can you help me clarify this better? So they much more implement questions like. So for instance, I have a Mercedes 190 diesel from 1985.
My son drives in it and it's old, it's more than 40 years old and there's stuff wrong with it. He hears something, he doesn't know what it is. When I would hear it, I would most likely understand where that noise is coming from and where we need to look at to fix it.
He has found a way to use the thing he hears and then turn that into how do I question this? Into an LLM, which gives me context here, this, this and that, and then give me a sort of answer. So I understand the part that my dad automatically knows and he's using that.
[00:41:28] Speaker B: Interesting.
That's really interesting that I like that. But the weights, the way that they're interacting with them, not just what they're asking, it's the way it's being done.
[00:41:37] Speaker A: Yeah, yeah. And I've explained them to. This is how it works. So the more context you give is going to help you get a better answer. But it's also not just about context. It is also, as you're building your context, you have assumptions in there and understand that those assumptions are there and they're not all necessarily bad, but they are there.
So instead of doing this long of a blurb and you know Have a full pair, have a full A four full of content.
You can start by the first paragraph, have it regurgitate what you just said and check for this. So I, I, I very actively teach my kids.
I mean, they're at ages where I don't do much teaching anymore because, you know, I'm an old stuffy dad, so whatever.
But I help them figure it out in a more logical way. And it's funny to see how they interpret that into their own, own world because Rohan, my son, he, and you've met him, he is, he's a highly creative person in the sense of photography, videography, playing music, writing music, all that sort of stuff. And it's very interesting to see how he wields AI to his benefit while staying fully creative and all that sort of stuff.
[00:43:04] Speaker B: So is he using AI to create music? Because there are music generators out there. So he's doing it himself.
[00:43:11] Speaker A: He builds it himself.
[00:43:12] Speaker B: Yeah, good.
[00:43:13] Speaker A: But he'll create a plugin for Logic Pro or DaVinci to help him in a certain way that the software can't if he needs it. Like, that's his.
Yeah, and it's fun to see how everybody uses it in a different way.
[00:43:33] Speaker B: I mean, I've been using it differently with my kid. As I said, he's only four and a half, but I'm still exposing him to LLMs and there's a couple of things we've done. So sometimes he draws characters, so I then take a picture of the character and I make it come to life in an image or a video. That's quite cool. I've been doing that stuff. Or sometimes, yeah, sometimes I go into something like Grok or Gemini and I go, and this is something I've been doing. He likes coloring in, loves coloring in. And I was buying coloring books and instead of paying like €5, he uses two pages and then gets bored. I'm like, I'll print out outlines of things. And then I realized that, well, LLMs can make outline versions of any image. So I'm doing things like a picture of him and then I'll make it an outline, print it out. He'll then color it in. But then he'll, I'll say, well, describe who you want to draw and he'll be like a rock star with like a fiery mohawk. I might go for it. And I'm typing a prompt and he creates a character which he then colors in. But then he also likes using my phone and I want him to play games, ones that at least make it you Know, not brain rotty kind of apps.
[00:44:35] Speaker A: Yeah.
[00:44:36] Speaker B: So I downloaded Tic Tac Toe Noughts and Crosses. So I downloaded it from the Play store. He plays two games and then he has to watch 45 seconds of ads. Me, the builder in me goes, nope, I'm making my own. So I vibe coded with Claude a PWA that not only has like Tic Tac Toe, I made eight games and I did it in two hours. Sat by a pool using Claude Dispatch with my laptop in a hotel room. And then I just. All I had to do is connect GitHub to pages. When I went in later, it's the only thing I did. And then all of a sudden my kid has got an app that he can play. No ads, so no influence of the outside world. I've controlled it and I've done a coloring in one. So I then took those outline drawings and then now I've made a PWA that looks at my hard drive, sees an image that it got yesterday, renames it to what it is, then uploads it to the pwa and now it's in my phone. And a writing one as well. I've done a writing one so he can write words and his own name and shapes and stuff. But though. So we're helping each other, so he's getting an app without any ads or influence. I'm learning quite a bit on how to deploy PWAs onto Cloudflare pages and stuff like that. And just designing something that requires a lot of UX for children because it's ux. I didn't realize you need a lot of UX work to go into something. The more simple that something is at the end, the more UX has to be involved in making all of those decisions. It is. And again, it made me think of a normal person.
Can't make it to the way that I'm. And I know that I'm no pro. Right? But I also know that my wife wouldn't have been able. She would have given up ages ago to do it or left. Left a bunch of foibles in it. But I'm like, no, no, no, that's. That needs to move here. This needs different logic. So I'm really having fun doing it. But again, like anything else, you can take it to work with you and go on. What I've learned from making this game is that I can do A, B and C during my workflow. And I think that's really good for all of us because we're all learning and they're Learning as well. So it's normal for them.
[00:46:32] Speaker A: Yeah, I mean, that's a very interesting part where you're a son, four and a half, very much in the beginning of his learning journey.
My youngest is 19, is in the middle of doing her Masters of European Languages and Culture.
And I ask her every now and then when she presents me something that she needs to work with or asks a question or just wants to share what she learned. Like, have you used AI with this in forming your thoughts? And this and this and that. And it's surprising how many times he says, already thought of this, don't need to do that.
Like it's so much part of their entire routine. And not for generating the essay, because she enjoys writing the essay too much to have AI do that.
Nor do. I don't think they can get away with it long term. And she understands that part. She chose this study with a very particular idea in mind. Right. So.
But, but to see them interact in this world so intentionally. But so part of this new technology, it's. I think it's fascinating and I'm probably most curious for her generation and. Right. The generation behind that. And how will they adapt? How. How do we.
Because, you know, the WordPress is dying. Yeah. No, it isn't. It's shifting, it's changing.
But what does WordPress as a publishing platform look like when they become of age, when they are entering their mature level and no longer feel like any content that they potentially publish should be in a walled garden? Because that is what Instagram, TikTok and whatever that is.
It's fascinating to see and it makes me very curious.
[00:48:29] Speaker B: It is because there's no answer.
And you open two questions. It's not just the how are you going to do it with WordPress, but where.
Where is it going to go? And is the web going to be like split into two? Cloudflare seems to be operating in the way that it's going towards the pay per use, pay per read, all of that kind of stuff. They've got the digital wallets coming in, all of those kind of things, and then you'll have the open web. And I think WordPress has a place in both of those places. You know that the, the platform is open source, but it doesn't mean that the content within it has to be.
So I think it's a good, it's a good place for WordPress to be. I. Interestingly, I. When Alain Schlesser, he used to work at Yoast as well. The best conversations with him, he's a genius. But I like. I don't know, maybe he likes me because I've got the. I'm not as technical as him, but I have these weird ideas as well which can again enable something that he may be doing. But we were talking a lot about the future of not just WordPress but websites in general. The Internet, the way we ingest websites, not just information.
And our website's dead. Not just WordPress, the way we see a website going to die. And I think of it as a yes or no.
Part of me thinks, well, you're not going to. Well, let's use commercial for example. We'll use a high end ticket item like a car. I think that you need a website, you need something visual, you need the tours, you need to do all of those things. But if I'm just looking for a pair of trainers or a jumper, noting that Google or whoever will know everything about me, my context, it'll know my fashion style, it'll have my pictures of what I've been wearing before, may choose things to match all of those things. Do I need a website for any of those things? No, I just need to see the shoe, I need to see the item. So maybe websites are going to be like. Think of you remember when phones, you'll remember when the phones first came out you had all of these fascias and you could. And there were people at stalls on the high street and you could get whatever you want fascia on your Nokia 33, whatever. And I think maybe websites are going to be like that, same stuff underneath, but looks different on the outside, personalized to you. So I may go on to nike.com and I'll see a different version of a website than you will.
But it's the same content, the same shoe. Maybe not the same shoe depending on our preferences and our shoe size and stock and all of that. But maybe I'll see a different version of Knight than you will and what it's outputting is an aggregate version of just information. And we have the facia.
That may be interesting, but again I see WordPress playing an even bigger part in that role. To be honest,
[00:51:03] Speaker A: I think what you're describing is put into better words than I currently have in my thoughts, but I'm roughly in the same direction in the sense that there's been a huge focus on data structure. Right.
The whole thing of SEO is provide structure that is easy for a human to navigate through, but that automatically means it's also easy for a machine to navigate through and understand the structure of the site as it is intended.
The next step is essentially what you're saying because we have the structure already and we mix that with your preference, whatever that preference is, whether that's on the visual side of things or how it's being said.
And one of the things I've done in my life is I've done a study of nlp Neuro linguistic programming. And one of the things it teaches you is that everybody works with a particular way of processing information.
So we have five senses and you have a primary sense. So meaning, if I'm hypervisual, I need to be spoken to in a visual way. If I'm very much in the auditory kind, I need to be spoken to in a particular way.
If we're using all those types of approaches, because I've given that in my profile into how we present the structured data.
And then essentially there is no content to be optimized, but there is a delivery system to be optimized.
[00:52:40] Speaker B: I think that's going into that now. We're going into that direction now. Artifacts are a good way of, say, of using that. You can now do a search on Google and it will produce an image for you. And I don't know if you've heard of Andre Karpathy. He was the co founder of OpenAI, is now at Anthropic. I love his stuff. Every time he writes, I've got to read his long, long articles. But he talks a lot about NLP and the way in which input and output in LLMs are going to be different because the output's for the human more so than the agent, but the input's going to be for the agent, not the human. So we need to think about that. But like you said, data is.
I've been talking about data integrity for the last year and how important that is and how things like OKF and ARD may be solutions for that. Turning things into markdown so an agent can read it but still output it in a visual manner. So for us, and again, the visual manner will be different for you and I because like you say, NLP will make us relate to things in different ways visually. I may like a pie chart, you like a bar chart, maybe. As simple as that.
[00:53:42] Speaker A: I think that's a perfect example of the simplest thing where it can start. Like if I look at all the RSS readers out there that allow you to not only just process the RSS but to make it, make the text look like you like to see it. Like, do you like serif? Then use serif you like a weird handwritten style font, you can, it's like just pick any of the Google fonts and it will render it in that style. Right. You like a color, you like flowery stuff? Sure. I don't have a flowery blog, but if you'd like to process my blog in a flowery way, you can.
That's an interesting movement and I think we'll see that happen sooner rather than later, because we are, we're essentially, certainly with the, the wave that is Cloudflare really hammering down on the agent, on the AI facilitating side of things with regards to whatever we publish.
It's clear as day that that's going to move in, in a more personal direction.
[00:54:51] Speaker B: I feel like cloud flare is like the dea, they're trying to, they're trying to win the war on drugs. But yeah, there's. Nobody told them that they'll never win the war on drugs. There are people more passionate than him, than the, than the DEA or FBI who want to, who want to continue trading the drugs than they want to stop it forever.
[00:55:09] Speaker A: So I don't think they, I think they understand this.
So I work with Cloudflare and I have some inside information and it essentially boils down to they have chosen their path as this is what we're going into and Agent first is the thing we're going to focus on.
I'm tldring very much in line with this conversation, but it's essentially understanding that this is the direction, not necessarily the war that needs to be won or a battle or whatever. This is inevitable, that somebody needs to focus on this and they've taken up that role. If you've been paying attention for the last, call it 18 months, maybe two years of cloudflare diving much, much more into all things AI.
It's inescapable that they position themselves to be the translator of data, meaning these artifacts that you, you referenced, there's, there's ways to do that all fully at the edge on Cloudflare's layer and your WordPress site will be the thing that process the data.
In a sense, this is what I want to publish. But then from there, combine that with your, I don't know, eight different types of ICPs that you have defined, and every single ICP that sees that page sees a different rendered version because we have exactly the information we need or the language we understand that you need.
And I've already been part of a project where this is fully implemented, where the ICPs have been defined, where the different types of roles of the various ICPs have been defined and you get a matrix of 30 plus different versions of ICPs and we're talking to you, understanding who you are as best we can. But we also ask you with a little pop up if you're wanting to share a little bit more information. With what intent are you looking at this site?
Yes, that is the future.
[00:57:10] Speaker B: It is, it's going to be.
This is what excites me, to be honest with you. So all the doom and gloom stuff like we're all going to die by 2031. Well, okay, fine, we're all going to die, but if we don't. Right, let's think of a backup plan of we're not all going to die and actually nothing. I don't know, again, it feels a bit like, you know, Y2K. You know, everyone was gonna die at Y2K. Nothing happened. Like not even a fax machine broke down. Like, I don't know. I know we did some preparation for it on the maths where it was. We only considered 100 years, not every single year. Other than that, I think, I don't know. God help the people who are celebrating at 9 the year 9999 and have to deal with that all over again, don't they?
[00:57:53] Speaker A: Yeah, yeah. I was, I was working at a large insurance company for the Y2K project.
It was a year of prepping.
Every single machine was tested.
There were some minor issues, nothing major, nothing that even remotely in hindsight warranted the attention. But you know, if you're an insurance company, you deal with money and you take zero risk. So yeah, that was an interesting point project to be part of, certainly in hindsight, I bet.
[00:58:23] Speaker B: And it would have been a nightmare for me because like I said at the beginning, I'm not into risk and regulation. I just say stuff like, can you not just turn it off and on again?
[00:58:32] Speaker A: Yeah, yeah, no, it's, it's, I think the, the, the, the cautionary tale there is.
We understand something is changing because it's it. We understand that this is a limitation in the hardware slash software, it's in the BIOS mostly of your machine and it might become a problem. All right, so it might become a problem. Then probably good for us to start testing those scenarios.
So that was a very clear.
It was a literal second in time where, you know, stuff could have gone really, really bad and it turned out it was okay. We are now in a similar time, I think. But it's not a fixed second in time, it's a fixed Period in time where we're moving from this and we're testing our way to the next version, mitigating the problem. We kind of don't admit it's there because it's evolution, it's new.
[00:59:32] Speaker B: Yay.
[00:59:32] Speaker A: And I'm a sucker for new technology. So, you know, I'm the last to complain here but I think it's, in many ways it feels very similar to me.
[00:59:43] Speaker B: Yeah. And I think because we were maybe technically minded at the beginning and we were maybe more advanced or more forward thinking when the Internet came out, we're going to take this with the same attitude, you know, open mindedness, opportunity. You know, I don't remember seeing on the news when the Internet came out about like the travesty and the, and the general calamity of, you know, the paper industry.
Did you like? No, it was all about new opportunity. I don't know, maybe again I wasn't noticing because I may have been a little bit younger in a kid.
No, it wasn't like that. They were like, no paper's going to be around for a while and all of that. I don't know. I think the ones saying AI stop AI now I think are going to be enlightened in the next couple of years.
There's no going back, there's no phase. This is going to be part of normality as much as a phone in our pocket is normality right now.
That's just the way it is. And to get that information on that phone.
When the mobile phone came out, I remember snakes and splitting sms messages into 140 characters. And now I was talking to you about making an app by the pool.
It's crazy. That happened in under 20 years.
Then again, like you say, everything's happening so fast and you look back at everything so slow and we now look at history, you know, beyond 100 years going, how long did it take to make the wheel like that? It's just happening. It's just the trajectory is going up quite fast and we're just seeing it. But our kids will say, what were you doing?
You were dialing up to the Internet.
Yeah. And you plug something in, what's going on? But you know, this is again the fun that I'm here for because you know, we're technologically minded, we're technical, we're in this ecosystem. Things like WordPress have a role in the future as well.
Which is again exciting to be inside. Not just WordPress but any publisher. Anything on the Internet I think is going to be all game for the Next year some are going to be in trouble. Like I'm definitely making things that I could have been paying a subscription for, you know, so some SaaS stuff I think is going to have big problems. But again, they're going to adapt, they're going to evolve and they'll make whatever their product is better so we don't have to build it ourselves. I think that might be the thing. That's the thing we need to think about what is worth building yourself versus what is just keeping it to the SaaS companies because they've been all right so far.
[01:02:09] Speaker A: Yeah, I mean, I see that as well. And there's plenty of stuff where I go like Glide apps. That was fun as hell when that came out because you just hooked in your spreadsheet, your data layer and you had a UI to have you create a mobile app or a desktop app or tablet app, whatever, revolving around crud for that data.
I haven't looked at how Glide is doing. I know they started integrating AI very early on, but I can't imagine a bright future for them.
For the simple reason is the Pope. The people that like to tinker are going to very quickly understand that that same type of tinkering they do inside a Glide, they can do inside a Cloud or ChatGPT or whatever.
There's plenty of SaaS companies out there that are solving complex multi layered problems. I think they're going to be fine. You don't vibe code yourself into the fully maintaining applications like that very quickly. Like technically inclined people, sure.
Developers, sure. But Gen Q public? No, I don't, I don't see that happening.
[01:03:25] Speaker B: What about the people? The products will be there and it'll be great, but what about the people? So we've seen quite a few of these, I can't even remember off the top of my head the companies that have done them, but there's at least five and they have the same script. It starts with a we're doing great and I'll tell you why. And then it goes through and then, and here's all the AI advances that we're doing and we'll tell you how much it's benefiting the company.
And then at the end they go, oh, also we've let go 25% of the workforce.
[01:03:50] Speaker A: Yeah. And we're raising prices.
[01:03:51] Speaker B: That's great. Yes, exactly.
But also great news for everyone.
I don't know. But I don't know. I haven't heard of a few of them in the last few months, but there was definitely a phase in the first half of the year where it was happening every week. And there were big companies as well. Like meta had 10,000, 6,000, whatever it was is. So it's not like even you're safe in the big places really as a, as a human. So I guess it's where do you stand? And it's like you were saying with you and your kids, it's how do you use these things? Not what do you produce necessarily, it's what's the working out in the middle.
Really important. I think now for the kids, none of which are going to be watching this podcast probably until it's too late.
[01:04:33] Speaker A: I know at least my son will process this podcast, so there's that.
[01:04:38] Speaker B: So it's all on you, son. It's all on you. You. You gotta look after us.
[01:04:42] Speaker A: Yeah. Yeah. Please.
I wanna, I wanna, I say please because I was gonna make the joke that I, I want to sit on the beach and do nothing, but I, I don't. I really don't.
Maybe for 30 minutes or an hour, preferably watch the sun go down. And that's about it. I, I don't ever intend to stop working.
That's not how my brain works. But interesting times for sure.
[01:05:08] Speaker B: I do. I'd retire tomorrow, but I mean, not working is different than retiring. So I would just do more stuff like the Noughts and Crosses app. If I'm bored one day I'll build something. If it turns into something great. If it doesn't, yeah, whatever, I tried it. That's the life.
That's the life I'd want to live. But still. Yes. Everything you just described there, none of it had anything to do with digital.
Sit on a beach, watch a sunset. You know, these are the things that are important. And I don't know, when everyone's on the deathbed, they're not all talking about how they were happy that they got into digital trends.
That's not their love of life.
It's the sunsets, it's the family, it's the people, all of that stuff. Yeah.
[01:05:48] Speaker A: It's the memories. Making memories is essentially what it's all about. And I guess that's our escape of.
I mean, for me, certainly as a 50 plus person, having gone through the analog time, understanding that that's what we're in the middle of right now. That's not life, that's a facet of life.
[01:06:09] Speaker B: Yeah. I mean, there's even someone I know, Dixon Jones, I don't know if you've heard of him, is in the SEO world.
He made Something called Entity Map, which you should have a look at. Right. So he experiments with loads of stuff. He's over the age of 60. Sorry, Dixon, I'm sharing that. But he just invested, like, I don't know, nearly €5,000 in some machine terminal so he could power Hermes and have his own olama and all of his local alarms. I'm like, he's doing more stuff than I am already. And he's like, well, I've got to do something here because it's everywhere and I can't not do it. And he's in a fortunate position where he's at the end of his career and he knows that probably in 10, 15 years, you don't need to care about any of this. But right now he still does need to care, but he can also afford to play around with the shiny things, you know, So I think that's really cool that even I would like to think that I'm still going to be like. I'm going to be like him then and be still interested in new things and still messing around. And if something cool comes out of it and he gets. He gets some revenue from it, then again, brilliant. It pays for that sunset he can watch.
[01:07:15] Speaker A: Yeah, I think that's a good one to. To wrap this conversation up on. I think that's. That's the thing we all should be focusing on the sunsets and, and. And live through the incredible change that we are seeing today in the. The TLDR is. Is humans consuming data?
Because that's essentially what this is. But it's just that part. It's just consuming. There's so much more to do.
[01:07:46] Speaker B: There's so much. But that's good. It's going to keep us busy. Oh, yeah, that's. That's the great thing is that, like, whilst there may be jobs at risk, there's another four jobs that don't exist yet, with titles that we don't know about yet that are just probably going to be yours or mine in the next few years. You just don't. You just don't know.
[01:08:03] Speaker A: And there's a lot of trade jobs that are coming back in importance.
So even if you don't like this part of the work experience, there's plenty of opportunity to do other stuff. So cars still need fixing, plumbing still needs to happen. Electricity is only going to increase inside of home. So, yeah, again, no doom and gloom, just fun.
[01:08:26] Speaker B: No, no. And until then, we'll wait for LLMs to cure cancer, get rid of all this trash in the sea, you know, until then, you know, like I said, it's the world. I feel like AI is everywhere.
[01:08:37] Speaker A: Right.
[01:08:37] Speaker B: But to me, the world is pretty much the same as it was beforehand. Like, there's no noticeable huge economic. Like it's. Yeah, it's creating jobs, there's booms happening, but no more again than the Internet.
I don't know. But until, until they've done that. And then there'll be the weird day where they figure out that God may not exist. And that's going to be interesting to understand how, how humans would interpret that answer if, if that happened, because that's gonna, that play that. What does that mean for religion? What does that mean for religion if AI figures out that religion has been a story? I'm not claiming that it is for anyone who's religious out there. I'm just. I'm just saying that an LLM may do that. If it does. If it does, what do you do? What's the Pope going to do? Do you think the Pope's going to go, you caught me, you caught me. Here you go, I quit. Let's close the Vatican and turn it into apartments? It's just, I just. Those are the interesting things about, you know, the, the humanity and us as a species. What happens there?
[01:09:41] Speaker A: Yeah, if we get to that level,
[01:09:46] Speaker B: will we not.
[01:09:48] Speaker A: I don't know. I don't know if we actually will.
Partly because I'm not well versed enough to fully understand the consequence of that, because that is Godlike in itself.
But also there's a few things I've read over the past year where people with ridiculous high level of IQ, like 200 plus, and they are all in consensus that there is a God, for instance. So the higher our intellect goes, the more likely there seems to be. And I'm not claiming any version of God is the one. I'm saying there is a version of is what they are saying.
[01:10:29] Speaker B: Yeah,
[01:10:31] Speaker A: I don't see an LLM large language model coming to the point where it starts taking over creativity because it just has enough examples to be new creative. I struggle with, with.
[01:10:49] Speaker B: No, we're not in ASI yet.
No, we're not. We're not in asi. But maybe when we are, I don't know. I think Jimmy Carr said a really good one. I think it was on Joe Rogan, which this podcast is definitely, definitely better than that. And that he said that maybe the LLM is God, because if you think about all the characteristics of an LLM, it's Godlike. And I don't want to say what he said verbatim or anything, but it's worth getting that clip there somewhere. But he said that it's benevolence. It just agrees with things. It has a source of authority. All of those things that would be the characteristics of what a God would be. And if it's going to get to an ASI level, it's only going to get more important and complex and influential on the planet. So maybe, maybe there's. Maybe it will be a new God for people. Maybe it'll be a religion. There's new religions. They still happen now and again, you know, who know, who knows? Maybe we all have to bow down to the machines after all. That's why I still say please and thank you. Yeah, always say please and thank you. You don't know. They. They remember everything.
[01:11:45] Speaker A: Exactly, Exactly.
Well, I want to say please and. And thank you as well, Alex. Fun, fun conversation on the philosophical side of AI. Very much I've enjoyed it.
Thank you for being on.
[01:12:01] Speaker B: Thank you. It's been lots of fun. I love talking about this stuff, but. Yeah, about being optimistic as well, because I've heard enough of those. Heard enough of those. What are we going to do? Go for jobs that don't exist?
This was much nicer.
[01:12:13] Speaker A: I agree, I agree.
Thanks again.
[01:12:17] Speaker B: Cool. Thank you very much. Thanks for having me.
