2026 AI & Ecommerce Predictions with Jo & Max (Agentic Commerce, AI Wearables, The Death of Prompting and more)

Hello and welcome to the first episode of the new frontier in 2026. -Jew. -Whoo. How are you? Oh, God. I'm excited that it's 2026. Big year ahead, very exciting. And we're doing what is often people's favorite episode, which is our predictions episode. And before we get into it, I thought it would be fun to look back at our last year predictions, just to see if it's worth listening to us, if you're listening to us. It is always. Joe, you've got a lot of stuff right. So just running through what you said last year in the 2025 episode. -I have a lot of stuff. -Yeah, you predicted that energy infrastructure and AI data centers will become a major focus, which actually feels obvious now. Maybe it was less obvious back when we were doing these in January last year. And there's been one trillion in commitments across eight major AI structures, including Microsoft, OpenAI, XAI, etc. You predicted that Google would continue to lose market share and it did below 85% for the first time in a decade. Meanwhile, chat GPT increased three X from 300 million to 900 million. Before I go on, do you think that's still going to happen next year? Like, what would you predict now? No, oh, wow. Well, that in Google, 2025 was your year. No, I think Google are going to have they already had a massive comeback. And I think that there will be no stopping them now. Actually, one of my predictions, which I'm going to share in the AI Freakomers Newsletter today, is around the fact that I have a feeling that guys like OpenAI and Anthropic will have to stop sprawling across everything and focus on verticalizing just because Google already is sprawling, but they have the data and the infrastructure for that. So, yeah, no, Google is going to now basically reap the compounds of its many tentacles and AI excellence. Yeah, I'm with you there. I think Google is, well, you can see from the stock price, right? It's back on the up. It had a bit of a shock of law, but I totally agree. I think I will them off so many times. The amount of times I posted something like Google just died and now I'm like, if we had done this prediction six months ago, I think one of my predictions would have been around agentic browsers. I don't know if you have this one. Agentic browsers was all the hype six months ago. And I now think Google Chrome, Gemini on the right hand side installed and they're going across all your calendar, all your Gmail, that will be the agentic browser. It will be Gmail, slash Google Chrome, slash Google Docs. It's already all set up for it and I already use Gemini in the docs, what's the cheat, whatever they call it. And it's very good. So yeah, I think Renee's on some Google, but you were right. They did dip market share. So you were still right, but we're not putting that anymore. I love this as my favorite sentence when somebody says you were right. You're always right, Joe. Yes, I know. And then we both had fun ones which came out to be correct. So you predicted people would develop an emotional attachments to AI chat bots, one in five American adults now have had an intimate encounter with AI, according to a New York Times and MIIT study, which is why I use the print so deeply. Why would it? Yeah, 72% of US teenagers have used AI companions, at least once, which is huge. And 52% of them do it several times a month. We must say, interruption kids, go out, meet people, don't do this. It's not good. Don't do it. And I'm sure you've seen all these new stories about the guy who's in love with chat GPT, but he's got his wife and he's sitting there and he's checking his wife and's there. What about the woman that married her chat bots? No, no, she didn't actually happen. It just for me, it makes me sad. It just basically shows what our lonely world will live in and how difficult it is to create a genuine human connection. Now we're so connected technically, but we are so disconnected. Yeah, it's sad. You shouldn't be like that. The one I got right, these will all be new ones that came true, but I predicted AI translations for pets. So I like to do the world and off the wall once. I said that we'd have them 20, 25. And obviously it's not quite here, but by do the big Chinese AI company, far to patent in 2025 to decode animal sounds and it collects vocal sounds, behavioral patterns and body languages and all this kind of stuff and it's decoding it. So although it's not here yet, I think I get half a point because it's happening. It's coming. We're giving you the half a point max, don't you? There we go. So bring it to you. It's 0.5 to me, but hopefully someone will be doing it. You're going to have a comeback in 2026. Cool. Could you want to kick us off with your top AI/ecommerce prediction? Oh, all right. I don't have a top one. Actually, I haven't really ordered them in order of importance. Actually, these I'm going to go together because they are connected. So the first one is that by end of 2026, at least 30 to 40% of all social media content, whether it's on X, LinkedIn, TikTok, Instagram, it's going to be synthetic, basically AI generated or AI enhanced. So there doesn't need to be much more said about that. Just endless amount of AI slope happening. But I also think that because of this constant downpour of bad AI content, there will be a comeback of human-made content. And I think much more premium value will be placed on that, where that is going to be connected off-short of things like specific, smaller communities that are very focused on human-made content, human-made knowledge, human-made value, whatever, are going to become much more present because again, the mainstream is going to be so oversaturated with AI that people would want to go back to some sort of more genuine real connection. I feel like the phrase of the end of 2025 was, "Is this made by AI?" at least from social media point of view. And so I think that's one part of it. I think the reason why also human-made content is going to have a massive comeback and people would be willing to pay more is because actually the current model is unsustainable from just a point of view of information creation. So till now, obviously we had one way to monetize, let's say, content that you create online and publishers would pay for it and then they'll get advertising dollars from visitors or there will be some sort of subscription base. But now with large language models going and scraping all of that and then essentially regurgitating it in the results and the Google AI summary is bit. That doesn't give any actual value-nited publishers nor to the writers or to the creators. And so that model is going to be very much very broken very soon and so there needs to be another way, another incentive for essentially quality new content to be created. We can see it with Wikipedia dying as well. And so I feel like there will be a completely new wave of, okay, good, slow, like human-made things coming back, which is awesome. I think that's how it should be. Yeah, it is wild. One of our partners is called a marketing. The founder posted a photo on LinkedIn of him meeting Messian Lewis-Warris. And I watched that Tim and I'm like, was this real or is this AI? You just do it literally on tellers. No, he's like, no, that was real. I was like, oh, because it would be like, when you said that, I thought it was a real thing. It's like I've seen so many of those on LinkedIn and I'm like, it was, the poster was obviously written by AI, which, no, that's pretty if you're listening. So like he had this like poster, obviously, which you can just tell. And then the photo, and I was just like, my, he's a kind of guy who would meet Messian, but also it just felt, everything felt very AI about it. So it's really hard. And obviously when you think about this and like the political sphere, like, it's really impossible to know what's going on in the world anymore with all these. And I was the one that was likely to know Messian, excuse me. He's just well connected. He's just a well connected guy. He did this LinkedIn post, like top 10 things I learned from meeting Leonardo Messian. So for me, it would be a little strange to do AI generated, but again, the post text was obviously AI generated and the images you just can't tell anymore. So I was kind of like, has he got to ghost writers to write this thing in a bit strange? Under the actually me, Leonardo Messian, then go into chat, you can generate 10, these things I learned. It was either you just don't know anymore is the point I'm saying. But yeah, I agree. Do you have a, you have a good one coming up? Yeah, 2026 will be the year of personalization. And this is bridging both the e-commerce and the AI things we like to talk about on our podcast. But I think product services, everything will stop being general and start being very tailored to individuals. So the e-commerce one is obvious, right? You're going to have super personalized recommendations from Rufus or ChatGee PT or whatever, but also brought it in AI will have AI tutors which goes to each individual student's pace and capabilities and at health care in workout plans across the board. Every app we're using will be personalized and I think we're going to rip like people talk to that personization for a long time. I think 2026 will really be the year of everyone has a totally different experience on every app geared towards them. Interesting. Do you know what? Like on one hand, I think this is going to be awesome. On the other hand, it's going to be so fucking boring. I'll tell you why. I feel like it's going to be one of those Spotify recommended for you, playlist shit, where you just get being fed exactly the same, I don't know, brand 30 tracks that you've been brinsing for half a year and you just never discovery anything new. It's like your Spotify made few playlists because mine are great always. Whenever it does a shuffle, I'm like, wow, I like this new song you've just sent to me. I do discover new stuff, but I discover new stuff which sounds like the stuff I like. Yes, sometimes, but I feel my Spotify wrapped is I have a very broad and very eclectic taste and of course it does give me a selection sometimes of new tracks, but just generally let's say 70 to 80% of the tracks I have already listened to and I just feel that that's not what a good experience is. It obviously depends on niche on niche. I think when it comes to something like personalizing for health, you want that to be personalized to you. You don't want any randomness in that, but for like broader things, like for example, there is a personalization when it comes to anything like fashion or even like sport, I would say sometimes it's actually good to include some randomness and some actual design, not just what the algorithm things that will satisfy you because yeah. I do use ASOS and everybody buy anything online, but let's imagine I'm a big ASOS user or whatever. I think it'd be great if it would just show me stuff that it knows that I like and suit my, because everyone's got a unique body shape, right? There's no standardization, everyone's wider or taller or whatever. So I think it'd be such a good experience if it's like, okay, we know Max, he likes to wear this kind of style, like genuinely plain clothes and then like any patterns on anything. I care more about the materials, but I want to wear like a nice cashier or something if I'm going to buy some new and he just shows you plain stuff from nice materials. I think it'd be great. Why would you want to scroll through stuff that is not relevant to you? Well, I think it depends. When it comes to fashion, it all depends on what kind of fashion user you are. I'm not a big fashion person, I buy everything vintage, but when it comes to let's say, if you are pragmatic, Elon Musk, I put you in this kind of category like this very functional form shopper, then that makes sense. If you are someone who experiments or is interested in new trends, et cetera, then it doesn't quite work. But yeah, there would be always exceptions. But I think personalization should be based on modern issues and there should be an element of randomness and element of actually introducing completely new ideas that may be dictated by the creators versus by the user. Otherwise, we'll end up living in a very boring world. It's the same thing when you think about social media algorithms and what we are being constantly said in terms of content on social media. At some point, it gets really boring because, for example, the only reason why I use TikTok is to get AI content. But I'm fairly sure that if I get a few other niches, that will be interesting. But the algorithm just understands, "Okay, she just wants AI and so that's all it feeds me." And so it's the same principle. Yeah, before we move on quickly on this, I think bringing me back to the listeners or this podcast who care about ranking on chat, ePT and rufus and stuff. We've already seen this happen whereby rufus is totally personalizing the questions that are promptly, product page and obviously chat, ePT is getting more more personalized. So it makes it a lot, if you're a eCommerce seller or brand, it makes it a lot more challenging to actually know how you stand because you need to be appearing to your audiences. And if you're doing like a generic ranking of what's my visibility in chat, ePT or whatever, it may be totally different to actually the people that you want to see it or they see it. So it's going to be interesting in terms of on our side at Zoma standpoint, from like the product perspective, but I totally think this is going with the years, personalization, shopping and everything. But yeah, Jerry, do you want to take the next one? Yeah, absolutely. I think that's actually connected with personalization. I feel that 2026 is going to be the year where much more noise and much more prominence is going to come to anything to do with AI wearables and AI powered hardware. From a point of view that I think the adoption will be probably wider. I'm very curious to see what Apple does. They are so far behind, but I feel that they will have some sort of come back. But I feel that the really interesting piece and particularly for an eCommerce perspective is going to be all of the non-structured data that is going to be available through the use of AI wearables and hardware. We have something like the meta-glasses, the Snapchat glasses, Google is rolling out some glasses. Amazon is rolling out some glasses. People are betting on glasses hard. Obviously, open AI and Johnny AI are also developing some sort of wearable that is going to be available in 2026. I feel that of course the first way is going to be the early adopters. But I think as long as it's genuinely easy to use and easy to wear, I think more and more people will start wearing it. What that means is that there will be so much data at the tips of these AI companies and then therefore at the tips of let's say advertisers eCommerce sellers. If we are talking about something like Amazon, actually with something they already rolled out and I published about last week, which is Alexa. Now obviously we know about Alexa and Alexa Eiffel, but what they have now done is that they've given Alexa also an actual visual screen. So the interaction doesn't just happen through voice. I know you're going to talk about voice, but it doesn't just happen through voice. It happens through also visual. But the point is that every conversation we have, even if it's like very, let's say, very boring and banal, like now is a data source for Alexa and Amazon to know, okay, what's happening with our lives. Now, extra play that to let's say something like glasses. We are walking around, we have our whatever meta glasses and we are staring at a pair of shoes for slightly longer because we like them. Now meta knows that this is the particular type of shoe that we like. Guess what is going to start getting served on like your social media feed. So I think it's a really interesting point from the sort of the, just from the user experience, like how much more we are going to be connected with AI just through the enablements through hardware, but also the data point and how much these companies are going to know about us in a way that they had never had before. That scares me, but I think that's going to be an even more present thing. Yeah, I didn't know about this opening eye physical gadget. I just looked at it as you talking sounds very cool. So from what I can see, it's going to be like an iPod shuffle, like little block that you put in your pocket or whatever and it will record everything. And I've long thought one of these things to be useful and I'm sure many people listening and you Joe maybe have felt this, you go to a conference and you're having all these great conversations and you don't have your AI note taker with you. I can't remember everything. So I think certainly in a business environment, it would be so to have something that just quietly recorded everything that happened and then you can go back and go through the conversations and thought out next steps, maybe we've got AI agent. So that's very cool. I've wanted one of these for a long time and I think open AI's one will be an acceptable one. They've got quite good brand or Apple. Like if one of these two launched it, I feel people will become comfortable having that in their day to day lives. Open AI at Apple, it's down to you guys. Build me my confidence recorder please. Make the most out of it. Make your shoes together. Do something for Max. Alright, what's the next step? So this is a big one and my sister if she's listening is working on a startup in this space and I think it's going to be massive. So 2026, the year of interactive video and world models. So we already have the start to world models in 2025. People like marble from world labs and genie three from deep mind. But basically what this can do is generate an IA3D environment from a front and then you can walk around in this environment and it will remember it so it's consistent. So you can explore these world models and go back on yourself and it's all consistent. So obviously this is going to the obvious application is gaming. This is going to make gaming very fun. I'm not much of a gamer. But also I think how does this change like social media and YouTube and like how all of us maybe who don't sit down and game like I never game. I do spend a lot of time watching YouTube and a lot of time on social more than I wanted time on social media. So I think keen I didn't exactly know what it would look like. But when we have these world models, how does that change how we interact video get news like all of this stuff I think is a if you can rather be scrolling Twitter for five minutes on a break you can just jump into a world model and get some cool insights or whatever. Yeah, I think that's going to be a very interesting one. Yeah, 100%. So like actually to, I think we need to maybe explain a little bit the distinction between these world models that you're talking about. And also another thing that I actually am going to write about today, which is world models, which are the next new frontier, the new frontier of the next wave after large-language models. So obviously, I think world models from that point of view also are going to be huge. And I just want to add that because yeah, I can't. That's what I'm talking about. Yeah. Okay. Yeah, the new world models as in the new frontier, the next one, say, whatever you want to call it, of LLM's. Yeah. So these are beyond, generative beyond just text, but you entire, the entire world basically. Yeah. So I think they are. And just from a, from that point of view, to add on to your prediction, I think they're going to be really interesting from a, let's say, a video point of view, but I think what they're really going to revolutionize. Oh, my God, I hate that word. AI, I hate that word. So not word revolutionized, but they're going to completely change is the sort of the 4A into robotics, which scared a shit out of me. But I feel like this is going to be like a huge factor as to how quickly we're going to start seeing some really evolved, really sophisticated robotics, which then obviously starts disrupting so much more about our world and our jobs. We, I was looking at the video from figure three, which was like now three, four months old. And you can already see that we are at a place where actual, let's say, home robots are happening. And yeah, okay, they're not amazing. They're not like I robot, but they are already good. And that is like not even three years after the first sort of 4A into that. And so the more we develop these world models, which are essentially models that don't just learn from large like masses of text, but they are learning from everything like from a point of view of images, motion, physics. So like they have so many more data points and so much more variety in terms of like their learning capabilities. That in itself is going to evolve robotics that's going to evolve video like video realism. And actually one of my predictions, which I think tags on nice to yours is I think in 2026 we're also going to see much longer video AI video generation. So I feel we started three years ago with three seconds and we're like, whoa, and now we have several minutes, let's say, but I feel like towards the end of 2026 we're probably going to go past the 30 minute mark if I'm so bold. Yeah, I think it relates to like the AI slot one you start to have like it's going to be it's going to be an escapeable long videos and stuff. Some of them is funny you say this because actually some of it is quite good. So for example, I follow this YouTube channel where they take old paintings and they generate from these old paintings what it would really be like to live in the fourth century London or whatever or 17th century New York. And I love history. So I absolutely love these videos and they just they show you the painting and it turns into reality and they go it's been like related to this world model stuff and it's all related. So I think that's I think it's a good use case, but for that and it's interesting that I've already started to consume in my free time AI slot like it definitely is AI slot and who knows I hope it's historically accurate. I guess the only story actually probably is that they start with the painting and then they just hallucinate from there. But yeah, I think there's definitely going to be more and more of this and it's yeah, it's a crazy way the world is going. Yeah, I basically anything I see in the news or on social media, I always just fact check. Like I don't ever take anything anymore at face value just because I've also seen like even news outlets like taking like AI shit and posting it because yeah, humans are fallible and extra valuable nowadays. Now my experience and obviously not the famous person I've been in a few articles now and it amazed me like the journalists in general get like 50% right. So I have a conversation with them and then they write it up and they'll just get half of it wrong. And you can't edit it, you can't change it. It's like they're in their integrity and it and to be fair, it's better than people trust us more, which is written and you've got the journalist name and that will be the source of AI models in the future, not random reddit, as you said, it's all going to be generated. Social media Wikipedia, all of this stuff. But yeah, there's no good information anymore. It's it's scary. Okay, do you want to go? You want to go? Sure. I for another one. You can find another one. I don't think you know he's. He doesn't find. I'm going to go. I'm going to go on a slightly more positive note rather than getting into AI do stuff of mindset, here's autonomous scientific discovery. I think 2026 will be the year when models are good enough to self improve and teach themselves and they already do this. But basically good enough to set up entirely autonomous AI labs that will just do scientific research and experiments and will push your boundaries in physics, chemistry, biology, all of these different areas. So I think obviously. Yeah, depending on how you frame that, I have a scarier exciting, but in general, obviously sciences and neutral thing. And in general, progress is good for everyone. So I think that's a cool one that will start to hear more about in 26. Yeah, I think you're right. I think obviously also with the policies I'm seeing coming out from the US. That is definitely a focus. I think it's a yeah, I think if AI can help with that. Yeah, awesome. Please, that'll be great. But let's see, I think there's going to be a lot of also conflicts of interest in that. And in any case, there will be obviously winners and losers, but I think in just in general, the more AI is used for scientific discoveries and progress, the better. I have one, which is more about more practical from an e-commerce point of view. And I feel like it really tags along nicely with what you were just talking about journalism, which is I feel that digital PR will have a complete renaissance in 2026. What we are obviously seeing right now is more and more conversations happening about a yo how to optimize to be like to be recommended in AI search engines. And obviously this is something this is a feel that I feel like it's continuously developing and changing what say chat to be T sites today, maybe different in what is going to sit in several months. So what definitely is is visible is that obviously all of these large language models and AI like chat bots are looking for very viable sources of information and within that obviously a lot of the sort of reputable sources are normally digital outlets. And so from that point of view, well, I feel like PR and digital PR work like a sunsetting priority for a lot of marketing teams in the last let's say 10 years. I feel like now because of the importance and the connection of like how to optimize for these AI search engines. And so I feel that this is going to become a much more of a focus so like more more, I feel brands are going to invest in how to become like how to essentially increase their brands, bread of mentions across reputable sources. And so I feel like that there will be I already see it like a lot more platforms that offer essentially citing and quoting within like different media and I feel like that's going to be yeah I renewed focus for brand teams. And I feel like that's really a million percent. I couldn't agree with you more than that one and I would say watch this space for what what we do here, but I think you're totally right for AI visibility. And so no longer need to be rank number one on Google you know longer need to be trying push for black backlinks for what you do need is broad mentions across all sorts of different areas. And I think as you say I think Reddit and Wikipedia will fast become slot and these AI engines will be like we can't trust anything from Reddit because it's just 40% AI generated crap so like it's no longer. For Wikipedia is no longer like a source of truth that we can use the only remaining source of truth will be publications where you have a human journalist in its name and the folks like New York times and business insider and all these guys will become. How I recommend product aimers product base I totally agree. Something we're thinking on very deeply at a zoomer so yeah I'm with you on the for sure. All right what do you have next. Okay I'll do one this was actually one of your predictions last year and I think maybe you are a little bit too early last year but I think it's definitely going to come this year we really touched on it. And I think that's the reason why I think it's not a big deal is that we're going to be able to use this as a device of voice AI and I've given example of something we did as a company which is 360 AI feedback so we did 360 back at the end of 2025 and everyone would talk to an AI agent. and you would highlight the 4 to 10 people in the company that you wanted to get feedback from and it would send a little email and then those people would talk to the AI about your strengths and weaknesses and then it would anonymize everything and give you an output report of this is your feedback, your strengths, your weaknesses based on all of the conversations that your colleagues have had with AI and it was all AI voice generated and it was an amazing experience and honestly I think with AI on the one hand you're happier to be rude to it so if it's being annoying you're just like skip stop asking me this let's move on but on the other hand you do open up more I think than you would do a human and I'll give you another example Jack and Jill AI which is also someone I've been using which does a recruitment for voice so what you'll do is you'll describe to Jill the kind of candidate you're looking for and then Jack on the other hand is having conversations of all these candidates and job seekers and it will then match you based on your voice description of what you're looking for and what people are hunting for it will kind of match you guys so I've just given two examples of real world voice AI applications that I've used in the last month or two and I think we're going to see this exploding in a very real way in 26. Yeah I completely agree with you I think as yeah as I said last year I feel like there is so much power in voice so much and encoded spontaneous information and inside that is conveyed through voice like tonality through pauses through warmth of voice speed all of that and that is data that humans can't necessarily decode but AI can and so there's just so much more that that is going to come out of that in terms of use cases both from just normal let's say B2B use cases but then also when it comes to shopping consumer ads consumer apps and so forth I think that will be really interesting and actually that's another sort of prediction that I I can tag along to this is with this and memory becoming much better within AI chat bots I think like personalization going back to what we talked is going to explore even more and therefore the influence of these chat bots are going to have and with that also the explosion of consumer facing AI powered like memory enabled voice apps so I feel that this is just going to be such a high potential for both for like essentially developers and entrepreneurs but also for the consumers proven also by what we were talking about before in terms of humans getting more and more attached towards their AI companions I feel like there is so much connected with these AI systems becoming more let's call them human remembering more what you are telling them and then being able to adapt themselves ever more towards your preferences your mood your tone your context etc. Yes Gary do you have any one or do you have another one? I have I have a couple smaller ones so I feel that what's going to continue happening in 2026 is AI consolidation so what I feel is that obviously we we had the last three years like so many different AI apps coming out a lot of utility being built around and a lot of AI wrappers or let's say AI technology with a little bit of sprinkle but like essentially based on a particular popular model but I feel that now we're going to see much more mergers and acquisitions happening last like last week metam acquired manas and that is obviously it just shows that these large guys are now starting to see okay where can we invest rather than build internally I have a prediction that probably Apple will be the most aggressive from that point of view just because there's at least on the surface far behind I can be proved wrong as with Google who do they buy flexibility? I don't want to make these bets yet but for example you're Tim Kirk who you buying? I don't know I from Apple point of view I actually I feel like I need to research that better but I feel like people definitely are not seeing some of the really special apps like Hicksfield, Jens Park they so much really top notch apps being built that are still probably have so much growth and like still can be like snap for not very much and have stellar teens and stellar usp there is a lot more out there I would be really interested to see also what some of the X open AI guys do because there's just so much now spreading in terms of new labs opening up and like essentially how that's going to play around in terms of mergers in terms of acquisitions like what are they going to acquire like how are they going to position themselves? Is it going to be more verticalized but yeah my prediction for sure is that like the AI consolidation is going to become a much more of a present thing rather than building internally? Yeah I think that minus one is particularly exciting for those in the e-commerce industry because it was a big favorite and the one you mentioned there at Jens Park like I learned about Jens Park on this podcast we have with Sim earlier this year and now I and he's better than Chatty I will literally if I'm doing like a deep research piece of work I will put the same prompt into Jens Park and Chatty research and every time Jens Park is better it's quicker it's more in debt it's it's more accurate so yeah I agree if I was at full I think Jens Park is a good one to go for honestly I think Jens Park and Manus were the two most genuine agentic applications out there they're not really agents but they're close but only the top agents gone I'll jump into my second last prediction agentic shopping will 2026 will be the year of agentic shopping and specifically this means Rookbus oh Max come on this is not a prediction this is not a groundbreaking thing I yeah you say that but I remember recording the podcast early last year in January we did two podcasts we did one with Ritu and we did one with Andrew Bell and Wana and Danny and we were doing these roof predictions how much is Rufus gonna be in terms of search volume and I said 25% everyone else not to tune my own horn was below like 5 and 10% and it was 38% by the end of 2025 according to sensor tower data and we will be getting the founder of sensor tower on the podcast shortly but explosion in agentic shopping has been enormous in 25 and 2026 I think it's just going to take over yeah you may say that it's not a spicy prediction I think the spicy one is I think we will get rid of the search bar and Amazon will feel more like Changchipotene in the sense that it's you use Rufus you don't use the industry send this agree I that's my vet against you I don't think that's gonna happen at all okay we'll do a bet for a what do we have last time like a hot dog in a beer I'll bet you the whole dog in the beer at the next event that that happened by the end of 26 that whole dog was terrible so let's bet on just some sort of like delicious food all right is it my turn or do you want to elaborate anymore okay fine let's get okay okay I'd say it's like tags on with the AI Slop prediction but I feel like it's going to be it's going further so I feel that a lot of what we're seeing in terms of influencers or digital creators right now on platforms a lot of these people are going to start using AI clones to like deliver even more content maybe a month ago I wrote a whole article about how AI has really find social media algorithms and like how right now it's so much more about frequency and volume of posting versus just quality and sometimes it's just like a throw everything and then one thing sticks and I feel that more and more we're going to see like the crazy pro pro-reflate proliferation of just like posting frequency and more and more content regardless of what the quality is and I feel that AI like basically influencers and just generally like creators are going to start using more of this technology to just generate more and then I can see that from huh what are you going to be one of these I don't really generate any video content yet just because I'm a bit shy but I should but now I feel like with it really depends on who your audience is right my ambition is not to go viral and so my ambition is to like reach niche audience and to deliver quality so I feel that the trust you build with an audience is more important than the views but I feel like for let's say bigger medium-sized creators who like have let's say sponsors and they are trying to get that reach this is going to become a much more being and for brands And you can see already tech like 11 labs and Hey Jen being so good. You can already see to an extent what sort of took and do with your AI avatar. So I feel like it's just. I mean, I would I try and stay off Twitter, but there's a big Twitter trend going on, which is just unavoidable if you go on to Twitter. The only fans style influences will be saying like hey, rock, do this do that. And like it's just full of AI generated images of rock of themselves doing sexual things. Oh, it's massive. Yes, massive trend. And obviously it's click farming because they get all the impressions and they get paid for it. So in that kind of industry, it's it's definitely happening now already. Yeah, scary. Yeah, do you have any more? My last one. My last one and I saved my I think most exciting with last, which is I don't know how true this is, but I'm going to say it in the most exciting and groundbreaking way. Promoting will die in 2026. And what will be replaced is AI native apps where they observe what you're doing and they intervene proactively. So you don't type, you don't ask it would just monitor your behavior and start generating the output that you desire. How would they know about your context? I'm more thinking about this in terms of B to B SaaS apps. So for example, StreamYard stream yard will watch what I'm doing every week or every other week I go and I set up this podcast. And that interface would be, hey, StreamYard, set up my podcast on LinkedIn and invite Joe whatever. And the new version, the proactive version, would it would just say, do you want to do this with this branding and everything and click and it will just you just do it. So I think there's the manual software step, which is where StreamYard is today where you're clicking and setting everything up and it maybe takes it and take long maybe two, three minutes. But let's imagine this is for a more complex workflow like procurement, getting paid from Cooper or someone this is like really painful, really manual clunky old school software. And the next generation of it would be you type in, hey, please do my invoice and it will do a few and that's brilliant. And then the next one after that is it just understands what you want to do predicts where you're going and it will just pop up and be like, hey, do you want to send an invoice to X Y Z and you like, oh, brilliant. Exactly. You've worked it out. So yeah, I did, as I said, I think this is maybe 27. It's a bit of a futuristic one, but it's an interesting one, right? Yeah, for sure. I think from a B2B Sus perspective and having almost like an agent that mimics like frequently used, let's say actions or process or flows. And I think that's very realistic actually, but yeah, I think from thing as a general thing is not going to go away, I think it's probably just going to be hidden from the user and then yeah, I don't wonder at all. Would that mean that some more controls will have to be introduced in the old school way, just because users, they crave control not complete control, but they always want to be able to amend things. And then we see how there is going to be. I think, for example, Cooper's a good one. When I log into Cooper to get paid by invoice by one of my customers, I just want to get the invoice like all the crap that I have to do is really annoying. And I imagine there's applications like Canva where you do want to have a little bit of control and artistic license and there's other stuff like, I don't know, health care. And if I think about using the NHS app, a national health service app or curing or stuff, well, I don't want to spend any time in there. I just want it done and the better and faster can predict what I want to do and do it for me. Like great use of Gen AI and even better AI. Wow, what an episode. There was a lot of predictions. So I look forward to doing one of this year when I put it into Gen Spark and I say, hey, which of these predictions came through and give me some examples. Thanks, Gen Spark. Thanks, Gen Spark and great listeners can look forward to that next year. Anything on the personal journey you excited for before we log off? Yeah, actually in next week or Monday, I will be speaking at the main stage at affiliate summit west. So if anyone is close to Las Vegas or wants to come and learn about AI live with me, you are very welcome to come. I would love to meet you there. I'm not going to be there sadly, but I will be in Las Vegas for shop talk, which is also at the same time. It's profit. That's not official yet, but yes, I will be very cool. I'll see you there then. Yeah, cool that I've never been so excited and enjoyed in January. That's very fun. Yeah, awesome. All right, have a great 2026 everyone. Thanks guys. Bye.