RR001: This is Realities Remixed & big trends for 2026

Realities Remixed, formerly know as Cloud Realities, launches a new season exploring the intersection of people, culture, and tech. Hosts Dave Chapman, Esmee van de Giessen, and Rob Kernahan unpack 2026’s defining trends, from AI and sovereignty to adaptability and automation, offering fresh insight, candid reflections, and forward‑looking conversations shaping the year ahead.

 

TLDR

00:20 – Introduction of Realities Remixed

02:30 – Why the show evolved?

04:50 – Dig in with the team: Predictions for 2026

06:40 – Macro trends

13:00 – Sovereignty 
17:40 – Agentic AI
22:17 – Human–AI interaction
26:06 – Cloud trends
30:42 – AI scaling, domain‑specific models
35:03 – Adoption lag
39:34 – Physical AI
43:47 – Quantum computing
48:21 – Hardware acceleration
50:30 – Cybersecurity
52:38 – Season outlook

 

Hosts

Dave Chapman:  https://www.linkedin.com/in/chapmandr/

Esmee van de Giessen:  https://www.linkedin.com/in/esmeevandegiessen/

Rob Kernahan:  https://www.linkedin.com/in/rob-kernahan/

Production

Marcel van der Burg:  https://www.linkedin.com/in/marcel-vd-burg/

Dave Chapman:  https://www.linkedin.com/in/chapmandr/

 

Sound

Ben Corbett:  https://www.linkedin.com/in/ben-corbett-3b6a11135/

Louis Corbett:   https://www.linkedin.com/in/louis-corbett-087250264/

 

'Realities Remixed' is an original podcast from Capgemini

Creators and Guests

Host

Dave Chapman

Chief Cloud Evangelist with nearly 30 years of global experience in strategic development, transformation, program delivery, and operations, I bring a wealth of expertise to the world of cloud innovation. In addition to my professional expertise, I’m the creator and main host of the Cloud Realities podcast, where we explore the transformative power of cloud technology.

Host

Esmee van de Giessen

Principal Consultant Enterprise Transformation and Cloud Realities podcast host, bridges gaps to drive impactful change. With expertise in agile, value delivery, culture, and user adoption, she empowers teams and leaders to ensure technology enhances agility, resilience, and sustainable growth across ecosystems.

Host

Rob Kernahan

VP Chief Architect for Cloud and Cloud Realities podcast host, drives digital transformation by combining deep technical expertise with exceptional client engagement. Passionate about high-performance cultures, he leverages cloud and modern operating models to create low-friction, high-velocity environments that fuel business growth and empower people to thrive.

Producer

Marcel van der Burg

VP Global Marketing and producer of the Cloud Realities podcast, is a strategic marketing leader with 33+ years of experience. He drives global cloud marketing strategies, leveraging creativity, multi-channel expertise, and problem-solving to deliver impactful business growth in complex environments.

What is Realities Remixed?

From the award‑winning team behind Cloud Realities, Realities Remixed explores what happens when people, culture, industry, and technology collide, because real breakthroughs rarely come from technology alone.
In each episode, hosts Dave Chapman, Esmee van de Giessen, and Rob Kernahan dig beneath the surface of innovation to understand what’s truly shifting. They speak with leaders, practitioners, vendors and thinkers across industries to explore how technology reshapes human behaviour, organisational culture, and society, for better and sometimes for worse.

Web - https://www.capgemini.com/insights/research-library/cloud-realities-podcast/
Email - [e-mail]

(00:00.75)
Alright, are ready? Alright, edit point. First edit point of Reality is Remixed. Woohoo!

(00:19.618)
I'm Dave Chapman. I'm Esmee van de Giessen. I'm Rob Kernahan and this is Realities Remixed, an original podcast from Capgemini. And this week, a conversation show about why we're evolving cloud realities and the impacts of this year's top trends.

So, top trends episode, Rob, you always look forward to those? Always exciting, David. Always exciting. Get into it. Get into it, right into it. Come on, let's have a chat. A new show Esme to talk about. Yes, love it, love it. Yeah. It feels like a fresh start. I don't know if you have the same, but January also feels like, know, blank page. Now I'm going to start exercising more and et cetera, et cetera. All those good intentions. So, but also very excited to start the show. Yeah.

Yeah, so let's address that to start with. So episode one of Reality is Remixed, we're here in a new space. Why have we done this? So let's just quickly look at why we've done this. So one of the things that we loved about cloud realities was the accidental discovery of insight. wow. Where it surprised us or we didn't see something coming that might have come from a personal experience of somebody or it might have come from, you know, seeing what was going on in a different industry and then applying it to a new industry or a million different variants of that. And I think as we sat down to talk about the show at the end of last year, and we've been thinking about this for a little while because we felt that the conversation has naturally broadened over time, but we were trying to put our finger on you know, what we thought was good and what we were getting good feedback about and therefore how we could build the show more closely around a bigger breadth of conversation as well as kind of what are the ingredients of insight. And we got to four specific words, I think, Es. So what do we think are the four things that when you bring them together, either all four of them or a couple of them that get you to somewhere interesting? So I think along the way we came

(02:34.25)
In almost every episode, it ended up talking about trust and trust between people that build culture. And we were talking about tech most of the time in business, industries, or the ecosystem that we spend our days in every day. So I think people, culture, tech, society, the impact that it makes, it always comes together in every conversation that we've had. The interconnectedness is so strong and it's also like so complex that

Either way, we end up talking about these aspects and if we don't, then we miss out, I think, on a certain level or perspective that adds so much value. So it feels right, I think. We've discussed it, right? So which words are, you know, what is it? And I think these are the words that resonate best if you look at the show in the past, but especially also what we want to bring this season. Yeah, agreed. We spent quite a lot of time thinking about those dimensions, Rob. What about them?

has provided even an analytical architect insight in the past. Well, we always ended up getting there. So we'd start with a domain and we'd always expand out to where Ez was discussing just there. So it felt like a natural evolution to just take that ground with us and bring it into the conversation at the core. I quite enjoy that because the bit I love about the podcast is it's a very big learning thing for me.

and you get to meet such an amazing group of people. You do want to expand the conversation out to a realm that has a different focus and interest to it. So hopefully we'll get into a more nuanced and complex conversation, but one that explores a whole set of new dynamics and areas. Yeah, lovely. the aspect of journey and exploration is something I think that's resonated with all of us as part of this.

said this to people like offline and chatting to them about when they ask about making the show, think we all reflect that the hour conversation that we have in making these things on a weekly basis makes such a big difference. And I'd encourage anybody to try and find even just an hour a week to have an out of band conversation.

(04:53.804)
I mean the best example for me was that the continued interaction we had with Anders and he kind of popped my mind the first time I spoke to him and then it's been a proper running complicated thread in my head about simulation theory and it's just a good example of if I hadn't have met the man I would never have thought about it but he has actually properly expanded quite a considerable part of my thinking because of that so that was good. Right then, crystal ball time. Now Robert you've spoken of your love.

of prediction. Everybody loves a prediction. Make a fool of all of us. Yes. And the specificity you like in them. And I believe Esme Robb said once, he's the best at predicting these things. He's a level four architect. know, if he predicts He in Vegas last year. He's like, he's like, guys, you know what? I'm really looking forward to the new show. What I'm excited about.

It's like, get a chance to do predictions for 2026 and I am going to nail it. Do know those people on the internet who claim they're from the future and they're coming back telling us everything is going happen? Yes, like the guy in Japan. Yeah, where the society is missing, but he's still walking around. Oh no, he's in Paris. That's basically what I am. But I've been a bit more intelligent about it and kept it under wraps. Right. Did we actually check it in hindsight? What the level of, you know, what is your ratio?

(06:20.526)
After we did 2025 predictions, Rob came back. Don't you remember when he came back and he was all excited? He's like, guys, guys, I've just checked out my predictions for 2025 and I was bang on. Don't you remember that is when he said that? No, I must've, I don't know, checked out.

I can't remember. I feel that you're going to stitch me up here, Dave. can, I can, my Spidey senses are tingling danger, danger. Well, I was thinking Rob's prediction. Well, let's call it, let's have a crack at calling it trends for 2026. Now I think there's a bunch in this that, that we've seen coming for quite some time and

you might be right, I think, in how it plays out. But I think there's some nuances actually in terms of where the conversations are going that I think we'll try and reflect in this. And all joking aside, this set are going to be interesting to look back on at the end of 2026, because I think one of the things that I'm sensing already is the complexity of what's happening at the moment is winding up. Yes. Most notably,

Is AI doing what we want AI to do? And what are the nuances in that? And how do we drive out value from what we have already? There's still that fundamental productivity issue that we need to tackle. AI was supposed to be the savior, but actually there's some bumps on the old road, aren't there? I think there's I'll get into that, I think. So let's talk about that. Now we're going to try and split up, as we sort of normally do, into about six sections, these predictions.

and within each section we'll try and decompose it a little bit. let's start, I think, Rob, and maybe you can introduce this, but let's start with some of what we think some of the macro trends are that are going on around what we're about to talk about. So I think there's sort of three for me here in all of it. There's the first one, which is how do humans and technology get better? So human tech teaming.

(08:28.512)
So we need to bring technology and human fusion together, especially around productivity and sort of thinking about don't linearly scale, but how do you scale exponentially using technology, proper automation, not the stuff that you often see gets implemented. That's underpinned by a move around two other things, is the sort of- I like the way you just threw a bit of shade there at historical automation. No, is. But seriously, we need to get-

much better at automation at a much higher level. That's just got to be part of it. And obviously AI and agentics will get to us a bit that. Then there's the undercurrent of the rise of the conversation around sovereignty, which is a lot of people now discussing that. And then the move from sort of like an efficiency mindset where we don't buy the cheapest anymore, but actually we're engineering more for resilience and flexibility because we might need to adapt in the future more quickly. So there's this shift to...

I've got to change my mindset to how I build these systems. I've got to get better at bringing the human into them and getting the compute to work better with the human or without the human. And then there's that bit about the whole, might want my own in the future, dot, dot, dot, dot, knows? So that would be where I was at. That's what I think is going to be the conversation on the macro themes in 26. So ever the...

ever the optimist on designing for resilience, not efficiency there. It just has to be part of the consideration, I think, more and more. As you have the conversations with the people in the industry, it is definitely a shift in thinking and it's happening. And it'll play out and it'll probably play out over a number of years, but I think 2026 will be the year where it's taken very, very seriously. So for me, I think I've got a mildly different take, but it's from the same route, I think, which is about

efficiency to adaptability. we've probably, I think we've talked about adaption and the importance of adaption in a complex world quite a bit on the previous show. The world, unfortunately, is not getting any simpler, is it? It's getting more and more complex. So I think the shift to building not necessarily for like a day one state of anything, but building for

(10:41.006)
whether it's in your technical architectures or whether it's in your organisational shapes or your processes or how all of those things come together or your governance. It remains for me to be about building for adaption. and it goes right out of the stack, doesn't it? So the whole ecosystem has to think differently. And there's going to be a shift in the way it needs to operate. I we cannot continue to operate in the way we have for the last 10 years. Again, the tech changed over the last two. I think now the humans have to adapt to that. And I think

what you're saying is aligned to that view, which is a lot of things have got to change and we've got to change quite quickly. if I build onto that, I don't think it's all about the thinking. I think it's really about staying with the anxiety, learning to stay with it and sense what's going on and sense what's needed, which is also asking for an emotional and a cultural capacity that I think is the most difficult part of all.

I've not had the best day coming into this record and I'm yearning for a nice clean solution. And I think what you're saying is like that solution is not available to me. So live in the anxiety and suck it up. no, yeah, staying with anxiety and not let it take you over. you you respond immediately, but you're just, if you're able to sit with it just for a moment and then decide how you want to act.

I think that's a real art if you're able to do that. That's a true skill that leadership needs to build, especially coming here. when you take that anxiousness, you sent through the note, Dave, about Nadella saying that, about needing social permission to continue to behave in the way it did, was referring to AI. I think there's an amount of social anxiety that's building around it as well, which is natural to see what is this going to do? Because the impact is yet to be properly understood.

I think that'll start to play out as well. Yeah. Well, let's move on to one of the other sub themes that you introduced there, Rob, which is undeniable and an enormous thread, is sovereign and what a sovereign means. So why don't you just, you're doing a lot on this in your day job, I think. set the table for us at the beginning of 26 and as we embark into the new show, how should we frame up sovereign now?

(13:02.378)
It's a funny old thing because you can't predict the future. However, the world is changing the way it's behaving at the moment. And so people are starting to look at the importance of digital in our society and what it does for us. So we cannot live without technology. And do I trust those who deliver the technology into me? So we talked a bit at Ignite about cloud was great because it abstracted everything and made it simpler. But now sovereignty needs to understand that abstraction to make sure we can rely on it like a utility style service. And so

loads of countries around the world are starting to look at understanding about where their technology comes from and how they deploy it and what risk might underpin that. You know, in the event of an event, will that still operate because it runs our society, etc. And so I think this year it's a very sharp look at do we like our supply chains that serve us our technology and what do we need to do to adapt to maybe adjust them to make us a little bit more comfortable. The flip side to that, however, is

is there's lots of things that have to be done and you need the benefit of the best-in-class technology. So although you might perceive a risk underpinning the way you consume digital services and the supply chain that brings them to you, there's also a bit about you still want to maintain the best-in-class tech to be able to make sure you get the best digital outcomes. So it's a balance. Six, seven there on the camera about the...

Hey, I've still got it with the kids, hey, that's six, seven, I'm still in it. Apparently they've all moved on. Have they? I'm late to the party. I don't know what to, but just as you're getting comfortable with six, seven, they've moved on. so it's a sort of game of risk versus benefit, but I think we need to think a little bit more carefully about it. That's how I would position it. Yeah, we talked about it, didn't we, in one of the shows at Ignite, I think last year, about this notion of abstraction.

And did abstraction go too far or was the world at the point that that abstraction was designed a place where that felt more aligned versus where we are now, which is to your point, I think that says the quality of the technology's utility that we have got to is maintained, but we have to be smarter.

(15:17.674)
about the landscape, the physical landscape that sits underneath that. Yeah, that's a good way to phrase it is we just need to peek back inside the abstraction and make sure we're happy with it. Yeah. Yeah. Or not. And then we have to make a change. Indeed. Well, let's move on to the second category and you know, no prediction points for this one, Rob, but you seem to be suggesting that agentic AI might be part of the conversation in 2026. Just tick that box, Esme.

And yeah, what's the, can you just, isn't it just fancy RPA, David? I think that was the, that was the debate we had last year. I think we've come full circle on it. What did Scott call it? Like just batch jobs? Just fancy batch jobs. I think it's probably a bit more than that, isn't it? As much as I enjoyed the charisma that came out of Scott, I do think there's a bit more under it. Yes. I think, I think the technology is more sophisticated than that and that's kind of undeniable.

you know, the big conversation that's going on at the moment, which we might return to in this section a little bit about, is it delivering the promise? Which is, you know, it's classic trough of disillusionment stuff. We've got to the trough of disillusionment. It's called that, it? Yeah, the trough of disillusionment. After the hype, you hit the trough. That's right. It's the hype cycle stuff. We seem to have got to that very quickly, we? We just zoomed, didn't we? Overshot load of a pasta things on the line.

So we're definitely at that point and we're going like, is it any, you know, is it, is it the route to AGI? Is it the route to this or not? So even if we park like all of the, where we thought we might be going and maybe we're not going there, but what we have today is ultra powerful technology and it's, it's, it's ultra powerful in at least two ways. And this is a super high level view, but I see it in at least two ways. Where one is in terms of like experience and interface.

and completely revolutionizing the interface that you have with, you know, compute generally, whether that be search or information or whatever it might be. So there's one element of it in my box that looks like that. And then there's a second element, and this is to your point, think, Rob, which is the levels of automation that can be driven with this stuff are on a different level. And I think to quote Robert Kernahan in this episode, then that automation stuff that came before it.

(17:40.778)
We're at a different level, aren't we, of execution. That's first thought on it. So unpick that a bit and then I've got another thing to challenge us with. the one for me, and as I really want to hear your point on this, is software engineering is one of the places which has been hit very hard with an advancement in capability. there's crazy figures being put out there, 10x, 60x improvement in performance.

And Agile as a concept was the new hope for being able to develop software in a better way and it absolutely works. But now by the time you've completed your retrospective and got on through your sprint planning, the software is like three weeks ahead and it's already done it. So can Agile cope and how does it need to adapt to the levels of productivity that even we're just seeing in a software engineering? Well, I think you're touching upon what a INAB-

actually already does, it shows how immature most organizations already are. If you're really able to have those agents in line with the work that you're already doing, so if you're already quite mature, then you really utilize the power of those agents. And then you're also very aware, where does an agent belong and where doesn't it? Where is really the capacity of a developer, which is creativity, but also business impact and complexity of the systems in the organization that the developers really know.

and then use that agent in ways that they can work faster and smarter. But I think, yeah, I don't want to put out some numbers, but I think a lot of organizations are like far behind if we talk about agile doing yes, obviously, post-its, doing the retrospectives, but the really being agile, I think that's going to be the organizations that really utilize AI agents to a level that you really do see, they're, you know.

They can be great team members instead of just an assistant. So yeah, I don't think Agile is dead. I was looking forward to an even more defensive reaction. I was being provocative there and as you've been very political and how you've treaded the course there on that response. Very good. To be clear as you don't think that Agile is dead. No, I don't think Agile is dead, but because I truly believe it's an evolution that's never done.

(20:02.67)
But if you look at it from a program perspective or everyone now got their new role and title then yes, probably agile is done But then you look at it from a program perspective But if you see it as an evolution because we're all evolving hopefully or sometimes you just wonder are we really or are we? But you know in terms of organization doing agile it should be an evolution that's continuously adapting on the outside world and the inside world something you're never done the

It definitely exposes the fact or it will expose the fact that some implementations of Agile have been pretty shonky and bad and are just masking waterfall behind the scenes. I didn't want to go that far. Fair enough. To go a level deeper and I'm just trying to see how mad we can get Ez on this. But when you say Agile is dead, Yes.

What do mean? I mean that by the time the humans have connected and done everything that they need to do and decided the next sprint, the software has already done the next five for you and is already like miles ahead. The sort of three developers in a room with 20 agentic agents are creating stuff that more so maybe we just have to consider we can achieve more in a period of time and the human will have to talk less about the detail. I don't know, but it just feels like we've developed this system that's designed to go faster and the...

Technology is saying, we can go even faster. Can we cope? Is there a big assumption in it that we actually implement the technology well enough to do what we are suggesting? I'm not sure there's a history of that. That's probably as checkered a history as the implementation of Agile, isn't it? Well, even if the tech is on that level, then you're actually excluding them.

outside your agile organization, you see them as something separate because it was actually the same company that was able to put that technology and AI agents in place. So they're part of your agile capability, right? And it also makes me wonder what is agile to you? And I think that's the most common question and you know, it's a philosophy, nothing more to it, to be honest. And it has so many practices. So it's almost like process martial arts.

(22:17.806)
Over Christmas, managed to expunge the image of Chapman doing Tai Chi from my head, which disturbed me a lot in November and December. And you've just put it right back in there. Es. I tell you what I'll do, just especially for I've got to send a video of me doing Did you actually ask someone, can you please take this?

But yeah, video, I haven't done that, but I think I've got an audience here that are willing to have a look at it at least, Rob. I'm ringing HR. What? No, don't do that. He's going to put this behind a paid wall. You probably wouldn't get to it through our Zed scale. Should we move on? Right. Where I was going was going to be a bit different to that, which was, I want to return to the,

human agent interaction piece. And I want to talk about one of our big topics from last year, coming back around demystifying some of this, is anthropophasization. It still didn't get it right. I actually tried to write it down phonetically in my book. It's the P you miss in the middle. Puh. Puh. Puh. Puh. Puh. Puh. Puh. Puh. Puh. Puh. Puh. Puh. Puh. Puh. Puh.

(23:41.962)
especially some of the interface aspects of it, mimicking human traits is illusionary, it? So the technology can still be excellent, but we don't have to think about it as mimicking human traits. this is very, I was with a very, very clever person with lots of PhDs who's doing robotics and getting them to learn like proper learning, not programmed. And he said, there's a really interesting thing that goes on. So

The more you make a robot look like a human, the more uncomfortable the human interaction gets with that robot until the robot looks exactly like a human and then it's fine again. So there's this weird, there's this weird effect that says, uncanny valley. either looks like a robot or it looks like a human and the more you blur the lines in between, the more uncomfortable I actually become. So there's this obviously a human response and you can't help but anthropomorphize.

things. That's just what we do. Anthemorphize. See the way I got the pf-f-f-f-f

to be as if that robot butler was a human or you want that robot butler to look like a robot but you're just chatting to it. Like R2D2 is my favorite example of I want it to look like a robot but behave with a personality of a human and so I can have a bit of fun with it and a bit of, know, like that. don't want just robotic response. Did you claim the name robotler?

Robotler. that thing? We have to check copyrighted reality remixed productions. That's mint. I love that. Yeah. Robotler. With SAS, with extra SAS. And then you can have modules that add on about how sassy you want them to be. And then you get up in the morning and you can move some slider bars around and have some fun. A SAS shift. A SAS shift. I need a SAS shift. Moving on. Let's do some more predictions. OK. This next.

(26:06.85)
bucket of things. We've changed the name of the show and the focus of the show, but cloud remains an omnipresent technology. So Rob, summarize us the future of cloud. because of the things we've gone before with resilience and sovereign, I think we're going to see continued fragmentation in the market and also the divergence of the offering. for a long time, cloud came together and looked the same and now it's diverging again. So there's lots of offerings in the portfolio that are changing.

I think along with that fragmentation, we're also going to see other organizations attempt to try and catch up. So Neo clouds is the new phrase. And there are lots of people, especially from a European perspective, that are building things that are trying to get to sort of hyperscale capability. So those Neo clouds are going to take a couple of years to come in, but they're becoming quite fashionable to select as well because they're different, they're built.

closer to home, they're geo-local, that's all going to come together. So I see more players entering the market, but I also see the existing market fragmenting as well. it's going back to your point, it's getting harder and more complicated to manage because of the side effect of all the things that are going on.

Complicated or complex, would you say? No, yeah, no, in the Kinefin framework terms, it's complex because it's changing very quickly and you can't manage it all. So it's not like your thing building an oil rig. It's actually much more foggy. Fluid. Fluid. It's a good word, Dave. Fluid. I like it. So let's take it. Let's take that apart a little bit. The building of new stacks. So...

Give us a little sense of what you're seeing, what's emergent there, and are we seeing stack parity with some of the more familiar clouds? It's a funny thing about clouds. they look the same at the base level, i.e. the virtualized compute pay-as-you-go commercial models, sort of like functions and PaaS services and things like that. But they have nowhere near the level of sophistication that the big clouds have.

(28:23.778)
But actually what's coming out of that conversation is that it's good enough. So actually a lot of cloud functionality that people chase down is pretty basic still. And there's lots of whiz bits sticking on around it, but actually 8 % of what you need is pretty simple when it comes to cloud touch. An analogy has sprung to mind. Has it? on then. Naming no specific products. But there is, you know, some products on the market that are spreadsheets that have got

unbelievable density of functionality. And then there are maybe other spreadsheet products on the market that don't have that level of depth and unbelievable density of functionality. But for 99 % of the users, they're easier, straightforward and do everything you need them to do. There is of course an edge condition that says,

people who do a lot of spreadsheet in complex macros, blah, blah, blah. They might not get the full... Is that a fair analogy? Two things on that Dave. Yes, it is a very fair analogy. But my second point is I was hoping for a lot more in that analogy because you got me all excited and you went to spreadsheets. I'm going to have to put an official complaint in around your analogy, albeit it was perfectly accurate. thank you for that. mean, so our producer has been working busily in the background and he has gone and

has gone unchecked for us, the validity of us copyrighting the name Robbutler. Robbutler, and the results are? The results are excellent. brace yourself Esme. So the word Robbutler, and this is in bold, does not exist in standard English dictionaries. It is not recognized as a valid word and its existence is debated among language enthusiasts. Right, we're in.

We're going to make that a thing. That's 2026 for us. Rob Butler will be, that's it. I think, I think the fact that language enthusiasts are already all over this Rob. I don't know whether we're late to the party. Dr. Mike, what, what, how did you do the research on Rob Butler? Did it involve a quick Google? Come on. just a search on the internet.

(30:42.51)
So in depth and extensive. was both not surprised it doesn't exist, but I'm very surprised about the level of debate amongst languages. Yeah. Seems to be generating. That could be, that could be AI just making up stuff though. It does do that. doesn't do it. It doesn't do that. I've had some absolutely outstandingly hilarious results from AI when searching for something and it just gets it completely wrong and just goes off on one. And it's like,

Well, I can't believe that that works. on that note, let's move on to our next bucket of trends. we are back in the world of AI in a very timely way with the evolution of AI models. I know it's exciting. So I think we're seeing in this, if I'm right, Robert is predicting. I feel like I'm taking a lot of pressure on this one.

that everybody is going to get their own model and that scaling will continue. But to the point we maybe touched on a little bit earlier already, maybe we're seeing the beginning of some early conversation about...

the value, the trough of disillusionment. So Robert, share with us. right. So the first one is everyone gets their own is they're now at a point now where you can train domain specific models quite easily. So all the hyperscalers and such like have given us tooling where model creation is relatively straightforward. saw that at cloud event. we seeing many people do that? More and more for particular domains, especially in areas like pharmaceuticals and such like with drug discovery and whatnot.

That is happening and it's happening quite a lot. think it's often in the life sciences we forget how impactful AI has been as one of the domains that has been transformed by it. But to your point, a lot of organizations need to get much better at deploying it and deploying it at scale and to the point actually affecting business models because it's just not happening at the moment. There's still all the stats going in that...

(32:46.274)
Most CEOs say they're not seeing the value and most pocks don't get to production and all that stuff. It's the year of truth for AI26, I think, because if it doesn't happen in 26, people are going to get very disillusioned about its impact versus its cost. And in the final point, there's so many people in the race in the market and model development is going at such a pace, but the ability to get the efficacy of the models right will mean that some of the big players may lose out.

And in that some of the big names you might see might not be the people who win. so 2026 might see a few earth shattering moments for AI. would you be able to see who created the model if you didn't know? So if you just look at it and you would screen it, would you see which industry or whose view is behind that model? So there's a big debate about

provenance in models, where they've come from, how they've been trained, what they've been trained on, when it creates a result. There's a lot of science going into the sort of trying to get that better, but you're going to have to, yeah, going to have to manage the provenance of your model, the AI bill of materials, where it's come from, the data set, et cetera. And I think we're also going to see a rise in the rigor associated with how did this model come to life?

And then the second phase is, and when I use it, how do I know it's done what it's done? Because there's still this thing about it can create results and you're not quite sure why it did that, et cetera. So there's quite a lot of effort going into trying to produce that. Remember when AI, you'd ask it a question, it'd give an answer. Now it gives an answer and tells you all the links it used to give you your answer. So you can go and check it yourself. That, I mean, that's a very basic version of it, but there's going to be a lot more of that going on. So let's talk about one of the other threads underneath this, which is the scaling.

aspect. Let's pull a few things together. So I think we've got like scaling, which we've been talking about for probably a good year at least, maybe a little longer. And there's emergent pictures on this, but there is the Nadella stuff that came out of Davos at the end of January, where he talked, Rob's already referenced it, but he talked about social permission and social permission potentially being lost unless

(35:03.534)
increased value can be driven faster from the tool set. Now, what all of this reminded me of is the conversations we had last year on Cloud Reality is about the adoption lag. And there was a rapid, unbelievable rate of technology innovation that followed the release of GPT-3, almost unprecedented, but it was very, very hard to...

understand how you brought that together with your people and your culture to be able to create something new and something good and whether that was tech complexity investment, not really knowing how to do, whether it be the ethical concerns. But it seems to me that there is a hairball of things here maybe that connect to the challenge of scaling and moving fast with this. Do you hold it?

the same sort of way and I guess if the answer to that is yes, then how do we start to unpick that? See, it comes to the new co-old co-debate for me, which says you get to a point where it's just too much for an organization to be able to cope and you have to create a new one and transfer the business across. And I'm wondering if we're going to see more of those because you just go, it's all too hard. Should we start again? Or companies fight the fight to

get this level of technology into their core, meanwhile all the startups bypass them. The cycle time for a company is now down to 15 years on average. So, you know, back in the 20s, like 100 years ago, it like 70, 80 years. And you go, well, 15 years ain't long these days. and I think it's obviously technology that's driving that. But yeah, I wonder if it's just too much. You've had the...

Organizations struggling with agile, organizations struggling with modern technology, then you load AI on top and then you load on agentic and everything else. Just suddenly just go, how do you adapt? When we talk about stacks, it's usually the tech stack, right? So the infrastructure layers, platforms, tooling, etc. But the real work, I think, is the human stack. But then, know, who feels responsible? Who dares to stop the system? Who actually sees where things go wrong?

(37:14.39)
in the entire value chain or the entire chain and who can actually stay in that as I said before in that uncertainty without freezing immediately. That's what Gene Kim's book about isn't it? Why are in the winning organization about slowification which is step back look at how it's actually working and go forward. That man has yet again put his finger on it earlier than everyone else in his book hasn't he? Let's be fair.

we should pay homage to Mr. Kim as he predicted that this would happen and it's playing out exactly like he said and he's already come up with an answer to how to cope with it. yeah. I love that. I love that take. Okay. Physical. So we've talked, we talked a tremendous amount about... was doing the same thing. Me and Ed started singing the song in the background. 80s classic. 80s kids obviously. you should put that in there. Just like a small...

Um, so let's get physical. I mean, literally couldn't hold yourself back. It's too hard. It's too hard. Sorry. It's too hard. Brimming over with it. Anyway, so we talked a lot about organizational, sometimes ephemeral feeling technology and, the human condition in all of that. Uh, but there is also a significant going significant amount going on in, in the physical world of hardware and,

an impact in that world. So Rob, run us through the headlines of that. For years we've been looking at robots and things and they've been doing funky things and whatnot. So they've got the engineering, mechanical engineering to a point where the robot can do things like humans and complex tasks. Now we're getting into the world of robots are learning how to look at objects, adapt themselves to be able to pick up in complex ways. then they can do backflips. more importantly, a human can say, pick that box up and put it over there.

and they can now understand what that human meant and then they can undertake that action. So you're now getting to a position where the robot is able to operate in very complex environments that the human is well designed for and actually can then understand the natural language interface. And suddenly you get this move towards a workforce that can be more robotic. So again, automation, productivity, AI fusing with the mechanical engineering to make things that can be very effective for us.

(39:34.254)
It's obviously in its early days, but you're starting to see some companies create some quite remarkable outcomes based on AI and mechanical engineering. And that's before you get to the fact that these things, you sort of touched on it with the pick up the box thing, but the where things like GenAI also becomes really interesting with physical robots, of course, is at the interface layer and then being able to pause and have very natural

conversations is pretty much there now and that's only going to get significantly better over time, isn't it? I also think there's a really direct connect here between agentic technology and physical robots. Yeah, if you think about the automation, now we can spring out of the...

the digital world into the physical to do something back into the digital and you can automate an off-normal. So when I talk about raising the level of automation, it can go between digital and physical and back to digital much more effectively. So you're getting to this point where you can do some quite amazing things. And we've already seen it with like...

very advanced warehousing and automation there, but that's very programmatic and structured. And now this can be much more dynamic as well with the way it operates. they can do that adaptability that you talked about earlier, or it's better to cope with the adaptability than just have to work within a rigid framework. Robo-agile coaches, That's it. That's the future. Robuttlers, we should work on that. Those are ro-bleh. Yeah, you can put an entire production line. Here's an interesting one for you.

It's still a big debate is you're likely to end up with a robot boss in the future or an agenting boss or whatever. A lot of people will probably end up there. How would you feel about that? Hard no. Might simplify things though. me stuck in my... Marcel's happy with it, like... Push the button now. I'd struggle with that. I'm not going to lie. I struggle with it anyway.

(41:31.032)
But yeah, any direction or authority, Chappers likes to rage against it. Deep, deep, deep, deep issue with, but the, no, the, I, I, what immediately came to mind as you said that was AI on the borders. Yeah. Yeah. Yeah. The, conversation we had. Yeah. That's what immediately came in mind. Like the, the, the level of sophistication and nuance that would be required in an, in an agent leader of humans.

We're miles off that, we? Well, it's not that... So to get it to a level of sophistication that the human could sort of find palatable, I agree. I think, however, you know, the corporate world and capitalism will see something and probably try and implement it anyway. I sort of feel that we're going to get some very clunky attempts at it first. Oh, that's going to be interesting to watch. It is indeed. Moving quickly on.

A couple of other things that are significant and going on in this space now and are only going to be brought up, like quantum I think remains an interest of ours on the show principally because quantum is in production in the Google cloud. I'm not sure whether it is in the others yet. It could be. I'm just out of step with that. It's not generally available, but I think you can request access to it now.

That to me is one of the big unanswered questions, which is all of the stuff we've talked about thus far intersecting with quantum and what that means I think is pretty fascinating, isn't it? Well, it is. You've got to remember though that quantum is only good at certain problem sets. So it'll be a fusion of digital and quantum together. However, I like the way that your show notes on this are excellent ways written. Quantum, old problems solved fast. exactly. The problem of...

David, it's perfect that because it's like take something that scientists have been trying to do forever, model a molecule as simple as a fertilizer. Suddenly you can do that and come up with a really effective one and massively change something like farming. Much more effective, much more sustainable. And that has a really big impact for humanity having to feed more and more people.

(43:47.918)
That's just one example of where a problem we've been trying to tackle for ages will be done. However, you know what I heard is going to fix as well? The M25. M25 problem. Exactly. Come on, bring it, bring that on. I mean, that is, that doesn't need solving. You can model the entire traffic situation and then fix it straight away and change it. Instead of adding lanes that take like, you know, three years of

having to drive at 40 miles an hour through road works, there's got to be a better way, isn't there? Well, yeah. You know what would also help with that day? Flying cars, which isn't going to happen in 26. Certainly not good ones. I we've all seen the haphazard job of that. I'm sure we'll continue the investigation into flying cars throughout the new show. 2036 is what that's going to be. Another decade away, that.

Especially not, you're not getting the ones you want either. You're not getting the flying taxi from fifth element Dave that Bruce drives. It's not happening. Never say never. Never say never. I don't like to bound my innovation thinking in the way that you do. He's throwing something of my own back at me. you. I like to start with a blank page Robert and build creatively and then work out how the technology can fulfill those goals. Shall we move?

Phrase of the pod. Right. Data centers. Very quickly. Data centers and compute power. we've talked about sovereign and we've talked about how that's going to drive some new investment in physical data centers. We know that the chipsets are constantly getting more powerful, those sorts of things. What's notable?

in terms of compute in particular, very briefly. the mini data center, everybody gets their own. It's very hard to build a big data center and get... when you say mini data center, do you just mean like one of the old data centers? No, mean... Different, different. I mean compact, modular and easy to deploy, easy to get power to.

(45:59.468)
And so can move it around if you need to. So edge has been building and building and building. There's lots of applications for it now. So distributed compute will become more common. they're moving into that and underpinning that is the sophistication associated with the fabrication of that, the way you put the racks in and the controls. So there's huge amounts of innovation going into how you build them and how they work. We take all this wondrous compute for granted, but there's some people who are working very hard to make sure that chips and things keep up with the demand.

And remember, NVIDIA stood on stage and they had a backdrop of a rack that could carry more data than the entirety of the internet. And you just sort of go, the level of innovation going into just the basic compute structures that we use is incredible still. And people don't talk about it, it's like hats off to the engineers working in those spaces. They are building some really funky stuff. People do take for granted, I agree, when a chipset comes out and people go...

Well, it's four times faster than the last one. It's half the power consumption and you just take it for Yeah, whatever. That's quite a head of Moe's law in some cases. And I think the rise of AI amongst all of the things that it's done has really driven hardware thinking forward just because people are trying to deal with the power consumption issue in a way that's responsible and all of those sorts of things. you're right.

That's showing up in multiple places, including though, RAM being difficult to get on the consumer market. Oh yeah, because they all went, we're going to flog this to the AI data center. And then all the consumers went, hang on a minute, I still want a PC at home. Did you do that? Did you build your own PC back in the day? Do you still do that? No, I don't know anymore. I did do it. I did it to prove to myself I could do it. And then I did it for a few other people because I'm a bit nerdy.

But now I stopped doing it because I can't be bothered, I don't have enough time. yeah. Have you ever built a PCS? No, no. No, that's the same reason why I love Apple. I just want it to work frictionless. Agreed. I noticed the brilliant coming together on our podcast WhatsApp chat about how convenient all of that was. I noticed Robert didn't take part. No!

(48:21.292)
No, I just let that conversation flow much like the one we had today around the logistics of this podcast. I arrived at 40 messages and went, I don't know what's going on. But even Dr. Mike has come round to it and got himself a Mac. Dr. Mike. Yeah. him a couple of days to get it working. Sometimes suddenly I saw the light. yeah. How's it going so far? Yeah, it's okay.

Well, I'm glad we got his insight there. My word. mean, that was a bean out. It's a big shift there from Windows 95 to... Still the best moment in IT history where they all got up a stage and danced. love it.

There's a lot of sweat on that. Last subject, cyber. yes. So we have got a few things going in here. think there are both cyber itself, the world of AI in cyber and then potentially the world of, we might be getting to the world of quantum in cyber. Yeah, I think you've got AI has been used to hack, AI has been used to defend against the hacks.

There's again a big thing about preemptive cyber as well, trying to guess it's going to go wrong long before it went wrong, etc. So going up the stack, think that's a trend that's arriving as well. And then you've got the whole, what does quantum do? Loads of people haven't implemented post quantum encryption. There is a solution today that will defend against quantum computers breaking your encryption. so many people haven't introduced it. And you wonder if somebody's going to go, that quantum works now and it can break.

the internet and everybody then rushes to update their certificates and how it works. So there's a lot of effort that's going to suddenly arrive very quickly. Now I'm not going to say that's 2026, but I am saying that we might be starting to get the nervousness in 2026 about what happens with post-quantum encryption. I've put everyone to sleep there, haven't I? You're all looking at me with glazed response. Somebody out there, somebody listening, hopefully at least one person found that interesting. Maybe one.

(50:30.446)
Martial was. I saw Martial's face. That's brutal. That's our trends for the year, Rob bringing that to a magnificent crescendo, but let me just quickly review what we said and give you bit of a summarized version of all that. So a number of macro trends that are going on around the industry at the moment that we will see reflected in either what we need to do to respond.

or we'll see it reflected in the product sets that we're seeing. So we're seeing, you know, destabilization and a continued increased need to be adaptable. We're seeing the hybrid organization maybe move from being a conversation into being more like reality. And of course, underlying all of that, we're seeing sovereign become more and more central to the argument. Agentic, something that is not going away. It is...

going to probably move to scale adaption this year. And that might have impacts on things like process and methodology, as well as the obvious automation upsides. In cloud itself, we're seeing reflections of

sovereign being taken into the stacks and being taken into our hyperscalers and others are responding to different needs in the world today. We're seeing how that therefore is creating more dynamic platforms, mixed ecosystems and fragmentation being something that maybe needs to be managed. In the world of AI itself, we're seeing this personalization of LLMs as well as

a continued scaling argument, but a scaling argument that's maybe shifting its emphasis a little bit and people are beginning to get into that trough of disillusionment and really beginning to say, where's the value here? Where is the value? In the world of physical, this whole new concept that Rob has introduced to all of us today, which is the mini data center, going along with the prodigious continued rise of compute power, which gives us more for smaller and more sustainable.

(52:38.126)
And then physical AI and quantum being the sort of new generation of technologies that we'll see coming this year. And then finally, Rob's excellent point on cyber. Rob, do you say the cyber thing again? Go away, David. Go away. Right. So there you go. Predictions for 26. Hold Rob to it. And we might return at the end of the year to just have a look at how those are coming up. Or...

as I might suggest, conveniently skate over what actually happened and we'll never come back to this conversation to prove whether I was wrong or

(53:24.588)
So there we are, a new show, books more specifically on the nexus of people, culture, industry and tech, and how all of these things come together and can be mixed to create value and insight. As part of that, the other thing we're going to do is when you bring all of that together, not only is where the insight, but actually how value gets unlocked out of that. how do you take some of the sometimes, quite frankly, conceptual things that we talk about and make those things real and make them concrete.

and actually drive value out of them, whether that be for the industry or the organisation, whether it be for the human that's at the middle of that and how they take that forward. Hopefully, as you can hear, we're excited about the new show, what we're going to do and the journey we're going to go on. Really looking forward to you guys staying with us on that. And as always, we encourage you to let us know. So as we go forward on this journey, as we'll give you the contact details, as usual at the end of every show.

We would love to hear from you about it. Is it working? Would you like us to go in different areas? Would you like us to explore things in a different way? Have you got any questions for us or the show? Even sometimes our guests. Absolutely open mailbox, so do let us know.

Now, even on the new show, we are going to end it with asking what people are excited about doing next. We think it's a good question. We talked about whether we should change the question and we thought in the spirit of the show that we've always tried to do is to keep it forward looking. And we think that it generally gives us a good insight into the person that answered. So we're going to stick with it anyway. So for this, let's talk about the new show and what we're excited about.

with this new format, I'm actually very excited to dig a little bit deeper in the connected tissue of humans. We did that in the past year in the show, but I think now I feel like I'm on discovery. So I'm very excited about doing that, to unravel and to see what we can find somewhere in between all the myths and some magic, hopefully. I think the magic is there to be found, whether we find it

(55:37.632)
intentionally or accidentally. It's almost the accidental aspect of it that I generally find delightful where you uncover something that you weren't expecting. Rob? do you think? I'm going to go for wider horizons, more learning, know, continued conversations with people who have interesting things to say. And then back to Ez's point about moving up into the culture and the people a lot more as well, sort of social impact, et cetera.

It's fascinating what's going on in the world and we'll be able to explore it a bit more. So that is actually genuinely sort of intriguing from my perspective. I'm also curious to see if we can find some, you know, if we have some facts about each other that like, what? Are you serious? Now, hang on, hang on, we need to wait it out, right? Because the whole Tai Chi thing upset me, put me right off balance, right? So we're going to have to moderate...

such things because we're going to do a blacklist. Basically don't talk about any of these things. So I like the fact that Dave will tell us he likes to play Scrabble on a Friday and we will not go any further. I think Rob, you know it's challenged you and you're uncomfortable with it. I'm just going to leave it there. Quite a few things make me uncomfortable in life and that is definitely one of them. I mean, I'm in the same place. I like it when we freshen.

things up a bit and try to create a slightly different skew in the conversation. We've got some amazing guests lined up on this and they're always amazing value. I'm looking forward to trying to evolve those conversations into something that's more about the human that's at the middle of it. We've done that on occasion before and I really want to double down on getting closer to that because I think that becomes much more revealing about the leadership challenge and the journey that you have to go on to implement some of these things.

rather than the technology. And then yeah, I'm looking forward to our challenge, the, you know, how we're going to try and bring themes together maybe slightly differently than what we've done before, is, you know, kind of, it jolts you out of a comfort zone to have to try and get to a different place. That's always fun, I think.

(57:53.26)
All right. Thanks for that chat on trends. Hopefully that's given you something to have a bit of a think about as we emerge further into 2026. If you would like to discuss any of the issues on this week's show and how they might impact you and your business, please get in touch with us at realitiesremix.capgemini.com. all on LinkedIn. We'd love to hear from you, so feel free to connect and DM if you have questions for the show to tackle. And of course, please rate and subscribe to our podcast. It really helps us improve the show.

A huge thanks to our sound and editing visits Ben and Louis, our producer Marcel, our Robuttler in the making and of course to all our listeners. See you in another reality next week!