AI Just Rebuilt Your PC Ecosystem: The Rise of Affordable and AI PCs | Utilizing AI Episode 25

The PC market is seeing a major reset driven by the impact of smartphone technologies and the emergence of AI. This episode of "Utilizing AI" features Futurum Research Director and Practice Lead for Intelligent Devices, Olivier Blanchard, Solutional CEO, Scott Robohn, and myself discussing the new face of the PC. Welcome to "Utilizing AI," the podcast focused on practical applications of artificial intelligence from the Futurum Group.

Every Wednesday, we explore news and use cases of the ways in which AI is transforming enterprise IT and the industries it serves. I'm your host, Stephen Foskett, president of the Tech Field Day business unit here at the Futurum Group. Before we dive into this discussion, let's meet who's on the panel today.

Hi, I'm Olivier Blanchard. It's good to be back. I am a research director and practice lead with the Futurum Group, and my focus is intelligent devices.

So PCs, mobile, anything that's not bolted down to a data center, that's what I cover. So I think, hopefully today's topic will be right inside my wheelhouse. And I'm Scott Robohn.

I'm the CEO of Solutional, a consulting company in tech. I'm a co-founder of the Network Automation Forum, and for purposes for today, I'm an interested and aspiring gadfly in what's happening with hardware. Absolutely.

And man, we should call this the Olivier episode. And I've been really looking forward to this because one of the things that happens every time I get a chance to talk to Olivier, I learn more and more about this intelligent devices space because each of us has our own ecosystems. I will admit, I'm in the Apple world.

I have my iPhone, I have my Mac, I have my iPad, I have my AirPods. I don't like the watch. I know I'm the only one that doesn't, but I like real watches.

But the point is, I am firmly in the Apple space, and what I don't know about is what's going on outside the Apple space. And again, Olivier, you have given me so much understanding of what's happening in Chrome and Android and PCs and AI PCs and the different market segments. Let's kick things off with that.

Tell us a little bit about what the overall intelligence device market looks like. Yeah. Okay.

All right. Let me summarize this in 30 seconds. So AI, right?

AI is coming to devices. It's already in a lot of devices, has been for years. Honestly, I hesitate to call it AI.

It's more really sophisticated machine learning. But essentially, a lot of the functionality that you're used to when you're on a browser and you're talking to Gemini or ChatGPT or Siri or Alexa, instead of being mostly done in the cloud, like those workloads, those inference workloads, are happening somewhere else, and then you're experiencing them through your device. But the compute isn't really all happening on your device.

More and more of that compute is actually happening on the device. And there are some advantages to that. One of them is cost.

If the device is doing the workload, you don't have to pay for tokens. It doesn't use up energy and resources in a data center. In terms of latency, having an immediate response from your device that doesn't have to go through a network and then back.

Also is much better as an experience because it's the secret to really having interactions with an AI or an assistant or an agent that feel organic and natural, like you're having a conversation with a human being that doesn't have his brain miles and miles away that has to process stuff. And then also for security reasons, having a lot of your searches, your discussions, your interactions with assistants and agents happen on device, where your data can also be stored without having to be sent in the cloud. Your queries also being held on your device as opposed to having to be transmitted to a server somewhere and then back.

There are a lot of positive security implications there. So there's been a push with semiconductor vendors and their device partners to bring as much of that inference, that AI capability, to the device for all of these reasons, for better UX, better security, better experiences overall, and then just, I guess, a more federated approach to resource management for AI workloads. And so that's what you're seeing.

So mobile phones have already been able to do live translation, for instance, or transcription. Mm-hmm. They've been able to do stuff like this.

You've seen a lot of AI that you might not realize was AI at first, optimize your photos and your videos and optimize audio, and your devices, like your headphones, do noise canceling. That's also increasingly more AI-driven. And of course, with PCs, you have assistants and a lot of system optimizing capabilities on device that allow your computer to run all day without needing a charge, without overheating.

You're starting to see a lot of these agents and assistants be on device for the same reasons that I mentioned earlier, which is that you have more privacy, more immediacy. If for whatever reason you're traveling or there's a network issue and you can't find Wi-Fi, or there are too many people in the network at an airport terminal, for instance, then you can still continue to work. You don't lose that productivity because you've lost your connection because your computer is AI capable.

So you're seeing a lot of that. That's generally what's the big transformation that's happening, from a technology standpoint, not an economic standpoint, on the device front. Yeah.

And this transition is really driving changes in hardware, and that's really kind of the impetusFor this episode was a few weeks ago, we were talking, and we were pointing out the fact that the mobile phone technology and the laptop technology and everything, it's all kind of coming together. And of course, we're facing headwinds here in terms of manufacturing. So talk to us a little bit about the changing hardware.

Yeah. So what's, I think, more interesting right this minute, because this AI stuff has been happening for a few years, and it's been a little underwhelming, and we can go back to that in a moment if you're curious about why we're not further along with on-device AI. But the real structural shift right now is not about AI PCs versus not AI PCs.

It's really a restructuring of the architecture of the PC market. So AI PCs have created sort of like this new demand driver and this new category of hardware, with NPUs basically at the center of the boards that essentially handle a lot of the AI enablement in a PC. But what we've seen is the advent of these extremely advanced, very sophisticated developer PCs, that are basically sort of AI servers in a small form factor, whether it's a laptop form factor, like a gaming laptop, they tend to be a little beefy, or a desk side or desktop PC with tremendous AI capabilities that are really for people who are going to be testing models or even building models as they become more and more efficient and as the hardware becomes more high-performance.

So you have these very, very high-priced PCs that can be anywhere from $3,000 to $50,000 and up. And then below that, you have your traditional sort of Mac versus Windows PCs that have varying levels of AI capabilities within the processors. So you have some that are Copilot Plus, which is a category of AI PC compatible.

You have some that exceed the minimum requirements, and you have some that fall below it because they need to be in a certain price point. But essentially, most of the PCs being sold right now, whether it's a macOS or Windows, have some kind of built-in hardware capability. But what's really interesting to me is that we've been seeing now for the last year, extremely strong signals that pricing of PCs is going to go up.

The first wave was because of tariffs, and for the most part, PC vendors tried to absorb a lot of those costs. But we already knew that PC buyers, whether they're in the enterprise, in the SMB market, or in the consumer market, were bracing themselves for price increases due to the tariffs. The second wave that's about to hit is component prices going up and primarily memory, because memory vendors are being basically flooded with orders because of AI in data center and other applications.

" Yeah. " That's it. Yep.

So, the priority is essentially going to data center and just these big AI infrastructure build-outs, which leaves PC manufacturers and other device manufacturers, not just PCs, a little bit in a bind. But essentially, what it's doing is when you have that much demand and a crunch on supply, it increases prices, right? And so we know that PC prices are probably going to go up.

And the fear across the industry, whether you're a PC vendor or a semiconductor vendor who sells to PC vendors, is, oh, no, if the price goes up 10%, or the price of a PC goes up 20%, is there a one-to-one relationship to demand, right? Are we going to start seeing a 20% drop in demand in PC sales if the price of PCs goes up 20%? We actually just did a study for one of our clients on this.

I can't talk about the results specifically because it hasn't published yet. What I can tell you, it's the one-to-one relationship is not something to fear. It's not anywhere close.

And there are a lot of mitigating behaviors that will keep buyers, whether it's the enterprise, SMB, or consumers, engaged in the buying process. They'll still be buying PCs for the most part, just in different ways. And some of the things that they'll do is they'll consider buying PCs at lower price bands, or they'll consider perhaps knocking a few things down on their list of specs.

For instance, less storage or a smaller screen or things of that nature. They'll downgrade some of the specs of their PCs. And we know exactly what they'll downgrade and what they won't downgrade, which is also super interesting and useful for PC vendors.

But so right at this time, when PCs are about to become more expensive, we see this restructuring of the PC space. With these super expensive professional PCs at the top, the regular Windows, and Mac. But now we're seeing this new type, this new category of PC emerge in the, I would say, $400 to $700 price band, which is sort of the sweet spot.

We refer to this as mid-range. It's essentially about 55% of the consumer market and about 45% of the enterprise market. So it's a really important price segment to nail and to be able to deliver PCs at.

And what we're seeing in this price band is a sort of a revolution started by Apple. I'm not always an Apple fan these days, but I think that they miss on a lot of things. But I think that this was a brilliant stroke of genius.

Apple introduced the MacBook Neo a few weeks agoThe MacBook Neo, if you're not familiar, is a MacBook in the $400 to $700 price range, depending on how you're buying it and what configuration. And it's a MacBook that's powered by a mobile chip. It's one of their A-series chips, not an M-series chip.

And so essentially, it's kind of like an iPhone in a PC form factor with a giant battery and a big screen. And it's wonderful. I have one.

It's making Macs affordable for people who could not afford Macs before, especially those who have iPhones. Now you have this really nice parity between their PC and their iPhone. It makes everything really nice, and it completely just opens up that very important price band for Apple.

And on the other end of the fence, you have Chromebooks on the Android side, which are also undergoing a transformation, and it's sort of semi-public. I don't know if it's an open secret at this point, but Chromebooks are going to transform into essentially Android Books. The exact nomenclature hasn't been officially announced, but we kind of know where it's going.

But essentially, you have this parity of Android phones and Android PCs opposite the iPhones and the Mac, basically mobile PCs, at similar price bands, that $400 to $700 price band. And so it feels like at this moment, when price points are becoming a little bit stressed with PCs, you have Apple and Android making a play in the PC space at that price band that might actually solve that problem for everyone. And the big question for me is how Windows, how Microsoft will respond to that.

And I think that they'll put themselves also in that middle band. They'll be playing in that middle band, but they'll have to compete now against Apple and Android in a way that they haven't had to before, at least outside of the enterprise. I don't think mobile PCs will definitely be a big game changer for enterprise players, but in the consumer segment, now you're going to have these three categories that are playing against each other.

Anyway, I've monologued long enough, but essentially that's the layout, or the lay of the land, if you will, for the PC market right now. So let me offer a comment and then a question. The comment is, it is really interesting to see new innovation in the hardware space.

I think the last few years we've seen it on the back end in the whole GPU and other XPU discussion, and now we see this coming on full force into the laptop market, especially as we've all lived under the mantra of software eats the world for the last 10, 15 years, right? And it's a pendulum swing, right? As far as I know, all software runs on hardware somewhere, right?

And it's the system, and it's got to work together, and maybe there's a better Apple conversation there on a return to good design principles. We'll see where that goes. So this conversation has my attention.

Here's the question. " Where I see lots of people running really interesting experiments, running agentic workflows on Raspberry Pis. Right.

So Olivier, how real is this? And Scott, stop being skeptical. How do I really process this?

Is this selling edge computing all over again, or is there more to it this time? No, it's a little bit like that. I think it's a mixed bag.

And never doubt yourself. I think that being skeptical about this stuff is the healthy approach. Not being a Luddite necessarily, although for watches it's okay, watches and cars, but for tools that we use for this kind of stuff, I think that skepticism is healthy.

I think that we've over-indexed on the need for AI on PCs and on most devices. I think that AI or these machine learning capabilities provide two things. One, efficiencies that we wouldn't find otherwise, right?

So even in the sense that your AI PC is capable of managing itself so well and being so efficient with all types of processing of workloads that it can run all day on the same battery charge without overheating, I think that in and of itself is enough. You don't need to look for anything else, right? It's a really good system upgrade, independent of any other sort of front-of-house AI, like an assistant or an agent that you can interact with.

The second thing, though, is I would say that having the capability on PC to have some front-of-house AI, so again, agentic assistant type of prompting interactions, even sort of the face recognition, the voice-activated prompts, all of those things are just nice to have. It isn't a game changer necessarily, but even if it makes the experience of using a PC better, so it makes you more productive, it makes certain tasks easier, it takes a lot of troubleshooting out of the equation. Like for instance, before the show, we were complaining about your device is not connecting properly to a PC, right?

So you're logging onto your PC, you have a Zoom call, or you're recording a podcast, you have your favorite headphones in, and for some reason, they're just not connecting. That kind of stuff is where AIcan help users sort of troubleshoot or get around those things. So you don't have to get into a menu, get into your settings, go find the thing.

You can basically have a prompt on your screen that you can pop up and say, "Hey, my headphones aren't connecting," and it just understands the problem and fixes it for you. Again, if you're using agents or assistants or just any kind of AI product that's generally in the cloud to do some editing or some fact-checking of a report, or you're working in Excel and you're working through spreadsheets and you've really been used to using Claude or some assistant to help you manage all this stuff and build out databases and other products around it, having that on your PC is really, really helpful for security reasons, for productivity reasons. I travel a lot.

Sometimes my internet goes out in the middle of something. Having the ability to continue working is really helpful. " Yes.

I want that to be automatic. That's it. Yeah.

So, it's sort of like taking all of that friction out of the experience, right? So you don't have to be an expert at troubleshooting to be able to optimize your PC or continue working when some of your resources like bandwidth or your network is down. I think those things are important.

So to me, AI in PCs, and devices in general, is sort of more of an incremental, very organic mixed bag of improvements as opposed to looking for that one killer app that people were complaining about, right? Like last year, "Where's the killer AI PC app? " And it's sort of like, well, that's not how you should think about it.

And then the last thing I'll say is it's also important to understand that AI is everywhere. We want ubiquitous AI. And so ultimately it shouldn't matter to the user, to the PC user specifically, where the AI workload is taking place, right?

There are going to be AI workloads that it makes sense for the device to handle and AI workloads that it makes sense for a data center to handle. And ideally, the best user experience is when you can't tell which is which. Just everything that needs to be quick is quick, everything that needs to be secure is secure, everything that needs to be big and impressive is big and impressive, and it's just all one big flow of information and insights, and workloads.

So, I would caution not to look for big flashy AI capabilities on PCs, but to think of it more as a system of systems that is improved incrementally or in leaps and bounds as it were. Sometimes it just kind of jumps and plateaus over and over again by AI. I think that's the goal.

So I have to bring something up here, and as a Mac user and an Apple enthusiast, I suppose, or Apple... I don't know. One of the things that has me really scratching my head is that-- So first off, I will say that the MacBook Neo looks fantastic.

I haven't bought one, but I tried one, and frankly, it's a MacBook. It basically is a MacBook Air that costs half as much. It does have some compromised specs, but none that you would really care about.

It's just that good. And so it costs half as much, and that's amazing. But it's a Mac, and yet I am a regular iPad user, and I've really been scratching my head over something Apple has been doing on the iPad side, which is they've moved all the iPads away from the A-series chips to the M-series chips, which is the chips that are found in the laptops.

So essentially, my iPad Pro, 11-inch iPad Pro that I use every morning to read news and whatever, actually has a faster processor and more memory, well, maybe, than my laptop, my MacBook Pro, and that's weird because frankly, iPad OS can't make use of that very effectively. And yet Apple has not only done this for the Pro, but they've done this for the Air. In fact, even, I don't know if the Mini has the M-series chips yet.

I don't have one, but it seems like they're aggressively moving the iPad over to the Mac hardware, just as they're aggressively moving the iPhone hardware into the Mac. What is Apple doing? Do you have any insight or any thought about what the heck is happening with the iPad?

Well, they're about to have a new CEO, so maybe we'll see some changes there. I think what Mac is doing, first of all, it feels to me that some of these teams are more siloed than they ought to be for Mac, right? So they're all working on their own products, and it also feels like in order to protect the tablet piece of Apple's business, touchscreen versus not touchscreen has always been kind of one of the parameters there.

The PCs can't really have touchscreens. The tablets can. But I feel like what Apple is trying to do is build the best possible version of each product that they can.

And sometimes that focus on design and that laser focus on making the one device better sort of gets in the way of having a coherent ecosystem of devices. And the way that Apple justifies this is by saying, well, if you're using a MacBook Pro or a MacBook Air, those are completely different uses from using a tablet, for instance, right? It's a different form factor.

You're not necessarily a creator on one, but you might be a creator on the other. And so theyThey sort of adopt this posture of different devices for different uses and different users. " I think that might help explain why you're seeing that and that might be the cause of your frustration.

I just don't understand it, basically. iPad OS still sucks and still can't make use of this hardware. It just seems asinine that my iPad has Thunderbolt 4 and my new MacBook Neo might not.

I don't know. I guess it does, but yeah. Yeah.

Well, okay. So that part of it, we're going to be in complete agreement, and I don't have any insight into why Apple chooses some of the ports that they do. We used to have these fights, or not fights, but these complaints about dongles.

If you're using a Mac or any kind of Apple product, there should not be any dongles involved. That's always been silly. But I think Scott has something to say.

Yeah, no touch, no pens either, right, in the Apple ecosystem. Not well. It used to be the case.

There's an interesting full circle here where you started out by talking about the re-architecting of the laptop market, which implies intent, and I think you have too many competitive forces there to see real collaboration on that front. But here on the Apple conversation, you do have a set of devices that, per Steven's astute observations, don't all seem to line up, right? And now, with the leadership change and somebody with more of a design background coming to the helm there, maybe we will see some trimming of SKUs and more of a story arc for the different products that are coming together.

I have no inside information, but that seems like a very plausible set of next steps. Any large company that grows tends to develop silos, and I think, Olivier, that you made a comment about product teams being too siloed from each other. I think there's some opportunity for them to bring this back together and have those more focused product lines that capitalize on some of the really good stuff we've called out here.

Like the Neo, man, that's a big switch for them going into that mid-range part of the market, and that's going to be disruptive there. Yeah. And they've done a really good job of not devaluing the brand of MacBook by doing it this way and having the A-series chip instead of the M-series chip.

There's always a risk, and I think that Apple has probably struggled with this for the last 10 years. How do we create a MacBook for that price band, right? Because obviously the revenue is there, and we just make so much money, and also capture so much more of the market early, students, et cetera, and move them up into the M-series MacBook category fairly quickly.

So they've done it without devaluing the brand at all, and part of that is just really good hardware, really good design. And I was in a coffee shop a couple of weeks ago, and I saw this individual working on a tiny little MacBook, and I thought it was a MacBook Neo, and it actually wasn't. And so they've done a good job also of making the differences so subtle that- Sure ...

it doesn't look like it's a cheap MacBook. It looks like a real MacBook. And it's cute.

It's orange or yellow. It's cute. Yeah, we have the yellow one.

So what I would say, though, is I hope that with the change of leadership at Apple, that we return to a more sort of-- Obviously, it's going to be much more product-focused, and that's something that Apple desperately needs now. They've been incrementally improving for too long. They haven't really had a huge innovation win since the Apple Watch.

I think that the Vision Pro was a great prototype, but not really a real product. The MacBook Neo, honestly, is brilliant business, not necessarily brilliant innovation. But having someone at the helm who is going to look at the entire product's ecosystem and, more importantly, the Apple user experience as a whole, will probably start to make the Apple device ecosystem a lot more coherent.

And so a lot of these weird things that have developed over time with these siloed teams working on their own things and optimizing their own devices might start to feel a little bit more connected as opposed to disconnected. At least that's my hope. And I feel like with the new leadership, we'll also start seeing Apple innovate on hardware again in ways that are maybe less nerdy but more obvious to users.

Yeah. That would be nice. Yeah.

Yeah. And ultimately, I would love to see just a better understanding of sort of what we're doing here and where we're going here with the various ultraportable Apple devices. But then over on the Android side as well, I think that we need a better understanding there, too, because essentially, I've heard the same thing in public that you probably have heard more about, which is essentially that Chrome OS is headed towardAndroid.

And that makes me wonder, is there going to be a whole raft of Android PCs? Is there going to be a new range of Android tablets? How are those going to compete with the MacBook Neo?

How are they going to compete with the iPad? I'm not sure what that'll look like. I don't know if you have any insight there and what that means for AI.

I think for AI, it's part of the driving force behind this. So I think that if you have better ecosystem integration where the tools that you use on your phone and the tools that you use on your PC and the tools that you use on the tablet are all the same. And especially for Apple and Google, because they also have...

Well, Google has Chrome. And Gemini is just so tied into all of the software and increasingly into the hardware. What you're going to see is this more frictionless experience when you're switching from one device to another or when you're using two devices at the same time.

For instance, I watch my wife use her iPhone and her MacBook, even the Neo, and being able to just switch back and forth between one device and the other flawlessly is really a nice experience as opposed to my using my Android device and my Windows PC. There are some integrations, but they're not as smooth. So now having the capability of having an Android phone and an Android PC or an iPhone and a MacBook, even a MacBook Neo at a low price, at those low price points, I think is going to help a lot with the consistency of the AI experiences, the consistency of the workflows, moving from one device to another, even finding your information, file sharing, moving your pictures from one device to another, from one app to another.

It's going to become a lot more fluid, and AI is going to be the layer, the user experience layer and software layer that's going to connect those experiences much better together. So that's what I'm hoping to see. And so the one company that's getting squeezed in the middle a little bit because it doesn't really have a mobile platform is going to be Windows.

And so we'll see if Windows becomes more of a professional ecosystem and less of a consumer-facing ecosystem. We'll have to wait and see about that. But that's definitely some of the weird competitive dynamic that I'm interested in following.

Well, what if, I won't make any bold predictions here, and maybe this is another follow-on episode. To your point, Olivier, on the AI assistance within an OS helping me with my headphone connections and so forth. We've seen the drumbeat of the death of SaaS with AI platforms.

There could come a day where we actually see a disruption in the OS level across laptops and maybe even phones that does more than an assistant. It would be really interesting to see how that develops in the next couple of years. I think so.

We're starting to see the beginnings of that, where AI is just much more built into the core of the OS. And I'm also seeing some PC vendors starting to integrate their own assistants or agentic layers, or containers rather, into their PC. So you might be running a Windows PC, but there's another AI layer that's embedded in there that's basically like a little window that's always on your screen, that's essentially your prompt window, your hello window for a dialogue window for an agent.

So it's going to become much more about that. And I think that ultimately, apps that we touch and interact with will melt into the background- Mm ... and the agent will be the layer that we interact with as opposed to the app layer.

So yeah, for sure. Really interesting stuff, and I agree with you on a lot of this. One more thing that we didn't really get to touch on is the impact of companies like Perplexity and OpenAI also wanting a seat at that table.

" We do have to run. This one has gone a little bit long, but for good reason. Thank you so much.

Olivier, I know people are going to want to follow you and read more and listen more. Where can they connect with you? Where can they find you?

Well, the best place is, since we don't have a lot of time, is going to be X and LinkedIn. It's not super original. com, which is where I write most of my stuff.

I'm equally boring as Olivier. You can hit me on LinkedIn or X, Scott Robon, one handle there. Reach out anytime.

Excellent. And as for me, you will find me, of course, right here on "Utilizing AI" every Wednesday. You can also find me on LinkedIn.

That's where I do a lot of publishing on the Techstrong gang. And of course, I'm also going to be hosting AI Field Day, May 13th through 15th. Check those out.

com to learn a little bit more about that. And actually, just a brief mention here at the end. com.

It's another podcast from the Futurum Group, where we're going to be talking about the big cloud stuff that we've mentioned here, AI in the clouds, which is going to be a lot of fun. So check out "Utilizing Tech" to learn more about that. " If you enjoyed this discussion, please do subscribe on YouTube or your favorite podcast application, and also consider giving us a rating, a review, and maybe a comment.

This podcast is brought to you by the analysts and experts from the Futurum Group, where insights meet AI. ai, the "Utilizing AI" YouTube channel, or the Techstrong TV app on your favorite mobile device. Thanks for listening, and we'll catch you next week.