What's Hot in Tech in 2025 (with Mischa Dohler)

It's been three years in the making but neXt Curve made it happen! Leonard Lee finally had Mischa Dohler, VP of Emerging Technologies at Ericsson on the reThink Podcast to discuss what is hot in tech for 2025 on a whim. Unscripted, press record and away the two went to cover a wide range of tech topics that will matter in 2025 from 5G to AI. 

Mischa and Leonard stumble upon the follow topics:

• Introducing Mischa Dohler, musician, technologist, futurist (2:50)
• What's on Mischa's tech radar? (5:30)
• Remote health and emergency services (8:40)
• 5G's maturity and associated opportunities (11:05)
• What happened to the Metaverse and where is it going? (13:31)
• The unappreciated present and future of the smartphone (18:45)
• The XR + AI dynamic - catalytic for 5G? (21:20)
• What does Mischa think about the overall AI trend? (23:30)
• Who started the jacket thing, Mischa or Jensen? (25:50)
• Mischa's AAA pillars for telecom's future (29:03)
• Mischa's view on AI's impact - Alpha Fold (29:45)
• The economics of our AI future - dystopian or utopian? (30:29)
• The great opportunity for AI misuse (31:14)
• Trust Infrastructure: Network APIs and the role of Aduna (34:23)

Hit both Mischa and Leonard on LinkedIn and take part in their industry and tech insights. 

Please subscribe to our podcast which will be featured on the neXt Curve YouTube Channel. Check out the audio version on BuzzSprout - https://nextcurvepodcast.buzzsprout.com - or find us on your favorite Podcast platform.

Also, subscribe to the neXt Curve research portal at www.next-curve.com for the tech and industry insights that matter.

Chapters

• 2:50 Introducing Mischa Dohler, musician, technologist, futuristIntroducing Mischa Dohler, musician, technologist, futurist
• 5:30 What's on Mischa's tech radar?
• 8:40 Remote health and emergency services
• 11:05 5G's maturity and associated opportunities
• 13:31 What happened to the Metaverse and where is it going?
• 18:45 The unappreciated present and future of the smartphone
• 21:20 The XR + AI dynamic - catalytic for 5G?
• 23:30 What does Mischa think about the overall AI trend?
• 25:50 Who started the jacket thing, Mischa or Jensen?
• 29:03 Mischa's AAA pillars for telecom's future
• 29:45 Mischa's view on AI's impact - Alpha Fold
• 30:29 The economics of our AI future - dystopian or utopian?
• 31:14 The great opportunity for AI misuse
• 34:23 Trust Infrastructure: Network APIs and the role of Aduna

Transcript

0:06

Next curve.

Leonard Lee: 0:09

Hey everyone. Welcome to next curves. Rethink podcast. I'm Leonard Lee, executive analyst at next curve. And in this episode, I have a very, very special guest, Misha dollar of. Erickson, he is the VP of Emerging Technologies, joining me on this special episode. And we're going to talk about what's going to be hot in emerging tech in 2025. Misha, welcome. How are you doing, my friend? Oh, thanks,

Mischa Dohler: 0:42

buddy. And just so everybody knows, we have not scripted anything. This is going to be improv, like real semi musicians. Yeah, I'm really looking forward to it. All good. All good. I'm sitting probably 500 miles north of you here in Silicon Valley. And yeah, good to be on your show.

Leonard Lee: 0:59

Oh, absolutely. You know, you're long overdue. We talked about doing this. Like 3 years ago,

Mischa Dohler: 1:08

which is what it worked out because we haven't planned for it. We just said, let's do it. I think yes, they just stopped me the email and that's it today. When the call just do it,

Leonard Lee: 1:17

you know, yeah, so we will be doing it. we are going to be, having a crazy discussion. No doubt. have no idea what we're going to be talking about, but all I know it's going to be good. And before we get started, please remember to like share and comment on this episode and subscribe to the Rethink podcast here on YouTube and, or on Buzzsprout to take us on the road and on your jog, you know, all your. Personal activities where you have time to listen to us and obviously on your favorite podcast platform. And so with that, wow, I am. So, you know, you had me a little bit concerned earlier on when you didn't have your red jacket on. I was going, you know, my

Mischa Dohler: 2:01

students, when they saw me that my red jacket, my red shoes, they knew I was in a bad mood. Something was wrong, but no, I have it on for you. Of course. Yeah,

Leonard Lee: 2:09

absolutely. That would mean. You were having a bad morning. Yeah, crazy. So, you know, I always love chatting with you whenever we have a moment, at the various conferences and events where we do bump into each other to have a chat, but then also, offsite, just. Having time together to exchange notes or compare notes. And so I want to take this opportunity to see if we can capture some of that, the magic of our conversations and talk about the things that you think are going to be hot. 1st, let's do this. Why don't you explain to the entire universe? Who you are, what you do, and why we have so many great conversations every single time that we meet. So the reason we have great conversations

Mischa Dohler: 2:57

Is you just have to look at the background of Lynyrd because it's full of musical instruments. So we, I think we both want to become musicians and it somehow didn't work out on the way, but I'm glad it didn't because I can keep it as a hobby. So originally I want to become a concert pianist, actually ended up in tech, in, the kind of history, but still have been composing, published a few albums on Spotify. I did a big gig in LA in 2017 in front of 4, 000 people. When we launched with my record label, my fifth album, that was my biggest thing on stage. I need to compose my sixth one. And back then I said I should do it with AI, never found the time to do it. And now I really regret this actually, maybe I'll still do it, but anyway, so I do music. I love, I love also policy work. I do a lot in policy. I've worked with, I'm actually on the board of off com is our UK regulators spectrum board there. I'm on the FCC tuck as well. Technical advisory committee of one of them, sub subcommittee there. And, yeah, I love that. And again, we also like to get hands on, we involved here with the engineering and the, the executive teams on all things, how to integrate our 5g. gear into all the sexy things, which, Silicon Valley is cooking. And that's really, I have to say, because I see things which you will get as a consumer in two, three years time. So yeah, that's what I'm doing.

Leonard Lee: 4:18

Yeah. And it's fascinating to get your foresight into a lot of this because, it is. The next thing, right, that you are focusing on at, Erickson from, quite frankly, a commercialization perspective, It's not so far out that it's part of, let's say, like, 6G research, but, the way that at least the way that our conversations always turn out, it's like, okay, look, we have this emerging technology. We have this technology trend. what is his potential to actually manifest itself into a viable commercial application? Right? Yeah. So that's the sense I get. I know that you dabble and you have all these technologies on your radar. No doubt. 6G included it's that our conversations tend to have more of a near term horizon and that's why I always like getting your takes because. obviously, as you mentioned, you're working on things that, most people don't know about but, you have that sense of where things are going. So, what are those things that are really exciting you in this year? This year is going to be a pivotal year for a lot of things, even based on my research and what I'm seeing. And as you know, I cover a lot of stuff. It's going to be a wacky year, but I wanted to give you an opportunity to share with the next curve audience. What is it that's really piqued your interest and that you think is going to be the center of conversation?

Mischa Dohler: 5:46

They're kind of the usual strongholds. I'll bring you to that in a moment, but I'm gonna take you on a little side road because you asked me, what am I passionate about? the thing I'm very passionate about is all the things I do in health. So it's something I started at King's college in 2000, maybe 15, 14, 15, 16 in the, early 5g era in the, Rather than jumping on building codes and, protocols and standards, I actually went out to society and asked them, Hey, in 2020, we'll probably have a technology called 5g. It will have these kinds of capabilities. What would you do with that? What could you do with this? You cannot do today. So I work with artists, massive attack, which really eyeopening, with transport companies, but also with our hospital at Kings. And, I. Floated the idea of using 5g enabled telesurgery, which back then we got a lot of heat for because people didn't believe that this was even possible. Zoom forward to the year 24, 25. And it turns out we're doing this now. So I have, I'm working with some of the most talented surgeons on planet earth. I want to highlight my buddy Vipul Patel. He sits in Orlando. as part of Advent Health and he's doing a laparoscopic surgery, he has done now probably more remote surgeries than anybody else on this planet. And it's taking velocity. And, of course the majority will be done with fiber. There's no doubt about this, but 5g plays a very neat complimentary technology for a variety of reasons. The first one is, believe it or not, as we started to roll this out, a lot of the IT departments say, Hey, we've never seen that robotic surgery equipment. I don't want to run this on my really vetted ethernet rights or fiber. So, therefore, separate line is needed. 2nd argument. This is that actually from a cost point of view. It's much cheaper to have, let's say, with Verizon, Timo, or at and TA fixed wireless access SLA great enterprise slides. rather than having a fiber paying one and a half thousand, 2000, dollars a month, right? Yeah. So entry into that marks much better of hospitals. And then the third one is, a lot of hospitals don't have actually good fiber. And then it turns out, not all the fiber, even in reasonably connected hospitals is at the five, six nines of reliability, having a backup link, just in case the fiber goes out and it has happened to all of us, this is where 5g comes in, right? So I'm working currently with our crew here in, from cradle point and, the Ericsson forks and the wider industry with the FDA, lots of surgeons, medical societies, hospitals. Healthcare providers, super excited. This is really what, keeps me going at the moment.

Leonard Lee: 8:30

Yeah, you know what, and as you were describing sort of this discovery that you've done as you've dug deep into this use case is its applicability in a lot of emergency scenarios. You know, I keep thinking about the wildfires that are occurring in Los Angeles, what we saw with the hurricanes in North Carolina, Florida. Where the infrastructure went down and there could be actually essential need for wireless connectivity where you just simply wire connectivity is down. Right? And so how do you create an architecture, a distributed computing architecture that could support, let's say on site, surgical procedures, or at least assistance, some kind of capability that allows. talent and expertise to be deployed closer to where the need is, because, it's physical, right? It's a physical thing and having to ship people out of, an emergency. Situation cost money, it's time and to be able to bring that expertise over the air into a location where there is, there's probably, you know, you might not be doing the surgery, but there's assistive things that 5g can bring in those types of scenarios. So I don't know. It's just something that Popped into my head as you were you know

Mischa Dohler: 9:55

You know, actually if you think about specifically LA, you know Probably a lot of the fiber pops really but down and said the fiber even though it might be It protected underground. It's completely useless at the moment. So in 5g naturally has also suffered because we got the towers and all that. But, you know, a company like ours, Ericsson, we do have these emergency response containers. We fly them in with. Helicopters put down hours on their telco infrastructures in there for first responders, for people there. And I love your idea. I've thought about this, actually, to actually piggyback, let's say, medical care doesn't need necessarily need to be really at this point, surgery, but, just. Smart medical 5g connected care, internet connected happens to be 5g. that could be a really big thing. Yeah. Let me talk to Barry and my CEO is really very passionate about this as well. And I'm sure actually we did help the LA crisis, with things like this. So it's a good idea. Love it.

Leonard Lee: 10:52

Well, you know, it kind of ties in with some of the ambulatory things that we're hearing about as well. And that's being experimented quite, or at least it's being POC, if not, pilot in in the UK. Right? I know that there's a case study there as well. So, I'm glad that you, you tee things up with 5g, because I think. I'll tell you right now. I think it's just being detrimentally ignored. I think obviously there's a, another technology that's getting a lot of attention and a lot of funding. I think 5G is going to be important. Whether people, consider it or not, and regardless of the degree of quote, unquote, disappointment, we're at a point where, 5g is at an inflection point. It really is. and I think it's, detrimental to ignore. That state that the technology and its deployments are, I'm looking forward to the, resetting of that conversation. And I like some of the practical views that you're bringing, even though, you know, you're looking at the future in a sense. But it sounds like you've gone through some journeys here with these technologies, use cases where you're starting to surface what does value look like when you apply these technologies? And usually it's not what you hypothetically thought at the beginning, right? You start off with something and you go through the journey and then. Those value cases start to surface up.

Mischa Dohler: 12:12

people ask me why you go for this difficult use case. My personal take on that, some really smart people have tried before me to bring that smartness into the healthcare sector and somehow. It didn't really always succeed. I'm not saying it didn't at all, but we don't have a fully instrumented, smart hospital with full connectivity AI today. And I think one of the problems was really that we kind of trickled in different use cases, very dispersed. And then we also did a bottom up approach where we went by the I. T. Departments and they have a different visibility on that. So I thought, let me go the other way. Let me take the most difficult one because I thought if we cracked This 1, the rest will be easy peasy, then go via the healthcare system and the surgeons, because they put the pressure on the it department. So the digital transformation would be much easier. And so far it proved to be the right strategy.

Leonard Lee: 13:04

And so, you know, I want everyone who's listening and watching to know that this is the reason why he wears that red jacket. This gentleman is fire. You know, he takes on the tough stuff, you know, you have to be that bright beacon, showing everyone no, I love it. That's great. So, 1 of the topic areas. I really wanted to talk to you about. Was a verse, but really where are we going with it? If you're still arguing that it's like the big thing, then we have to have a conversation. But look, this is the thing, you know, me, man, I'm always fan of the technology and engineering that people do the great science. And research that goes into, to making these technologies possible. I'm not a big fan of the hype, you know that, right? I'm not for that. But what I know that you're very cognizant of, because you do this top down thing, which I think is a great approach, by the way, I think like disruption and real transformative use cases come from, expensive stuff, stuff that's tough. The easy stuff that tends to come about a lot slower. I mean, we saw that with iot, right? It moves at a snail's pace, but I want to get your thoughts on where did the metaverse hype? Take things and where are we landing? What's some of that residual stuff that the audience can start to think about that can exhibit value? Because I know you haven't given up on it. And if you did, I'd be really disappointed. But let's talk about that for a moment. Because honestly, I think there's a lot that was always there that has, you know, growing potential.

Mischa Dohler: 14:59

Yeah, I mean, we should probably, we should have asked Mark Zuckerberg to be with us on the call here. Right? So next time, let me give him a buzz. Oh, yeah, that'd be great. I mean, it's been, it clearly has been an overhyped and overloaded term. it means very different things for a lot of people. from different genres. the whole concept of that spatial internet, I don't think it's dead at all. And, I transited, I transformed, I grew from a, ambassador, into a true believer when I tried on the metal Orion glasses. I was one of the few lucky ones who could do that. And it has really blown my mind. So I don't want to. Give advertisement credit to any specific company. But I've seen the first time ever I have seen a sleek device sexy enough for me to walk around with it for a sustained period with great optical, and general technical capabilities. Right? So we're getting there. So the engineers, you give them a challenge. And they somehow managed to crack that. the jury is still open if people will actually adopt that. I personally always, try to believe that history loves to repeat itself. And we had one gadget which grew like, mad, which is the smartphone, right. I stopped with the iPhone, et cetera. So we have that data starting 2008. We got a super exponential growth there until now it's kind of, flattening off, but it's been a stellar story. And then we have the, let's say the smart watch. I like to use this as a lower bound, which is kind of a useful device, but not everybody's cup of tea because I don't wear it because it bothers me with playing the piano. So I'm not a smart watch guy at all, but you know, penetration has a 30 percent were much slower. So therefore, we could imagine that AR could be anything between these two and either scenario is a good scenario for us, a community use cases. Listen, I always like to listen to myself when I wear them the whole day and play games. I'm out of this age. I've just gone over 50. So it's downhill here from, but, you know,

Leonard Lee: 17:08

are you making fun of me? Come on, man. It's not that bad. It's not that great, but it's not

Mischa Dohler: 17:15

All right. I love you too, man. Anyway, stuff like navigation, having a call with your friend, et cetera. Or engaging, actually, what I think will be the even bigger use case with the, LLM, with a smart agent, very interesting. And Meta actually, you know, I'm not sure they truly believe this will really exponentially expand. So they've floated this idea and has been announced publicly, to actually have earbuds with a little camera, So, and I can imagine people wear this even more because it's not so intrusive. You have it all with you, the cameras is capturing like a 360. It understands, the spatial context advises you, tells you things. Hey, you know, the future is still wide open, but we're getting there. It's crucial. Spatial compute, whatever you call it, is not dead.

Leonard Lee: 18:00

All right, there, there you have it. Okay. So do I give my take? Do I have no, I think you're, you're tuning into a number of really good things. I think 1 of the caveats it boils down to silicon. Are we there with the semiconductor technology? And, can we get the power power efficiency? And, the density to support these form factors. And, 1 of the things I've noticed is that a lot of these devices are becoming actually, smartphone reliant. They're not standalone devices. And I think, with the Orion, 1 of the things that they didn't make quite clear, and many people didn't talk about is that you do have that puck. Right. And so, oddly, I like that you brought up the smartphone and how it went universal, right? It's a universal device and it seems like it still continues to have a role. And I think it's 1 of the reasons why it's not a, not something that to be ignored the innovation that's happening is in how it's becoming increasingly multifunctional, but it. It's expressing itself in very boring ways that are actually very transformative or enabling other transformative, let's say, peripherals like smart glasses. Because I've tried. I was at CES. I tried a bunch there. Getting better. I have to say in terms of, any of these XR experiences, division pro by far destroys everybody. they're so far ahead. It's not even funny, but, I think in terms of the AR stuff. no, and the only reason why I say this is because I just see so many haters out there talking nonsense about the device that they never even tried, that I think we'll get there. I think we're going to have to start with simpler and that's what seems to be getting traction like metas Ray ban classes, but the use case and a lot of people talk about. Is the camera, even though they wanna put a lot of attention to the ar. When I talk to folks who, use that device, they talk about the the camera, taking videos and stuff, and having that first person view, one of the things I have talked about for quite some time, the immersive media areas is probably gonna be a pretty hot thing. And even with some of these smart glasses, one of the key use cases when I'm on a plane. I can either do productivity or I can watch a movie on a gigantic screen, as silly and trivial as that sounds, especially with these planes. He's getting small and smaller. If you're not flying first class like you, it's something that I think has growing.

Mischa Dohler: 20:51

but, you know, I'm with you on this actually, and we will probably see as we, have seen as big as with a smartwatch, right? So the power users. So, and that's what we are interested from an Ericsson point of view, what does it mean? And typically, you know, Pareto, Zipf's law kick in at some point. So we know that, from all the population using the device, about 20 percent will be power users. And the question now is how much power is power, right? Is it, will that be using that for 24 hours, 12 hours Three hours, because that really will influence the way how we need to dimension the uplink, and the downlink and the uplink. If you think about it, you have these glasses on and they're actually interestingly not streaming video at 60 FPS frames per seconds. They take, maybe once per second, five. Per second. So if you work at, let's say five, 500 per second, they need a much lower resolution. Interesting because the LLMs get away with recognizing objects at, 480, P resolution, which is really low. So you get something like 200 kilobits per second uplink roughly. but still, still, you have. 20 percent of population's power users, doing this for, let's say four hours, Verizon, Timo and AT& T need to start, seriously considering upgrades to get, we need more stuff, we need more upgrades and all that. So for us as an industry, I think it's an exciting time. but we still have uncertainty on, how many users will really pick those up.

Leonard Lee: 22:07

And then how quickly, I think also the architecture. Will evolve, and then to your point how does the data flow, but you're making a really interesting point about the, AI vision that, or, what folks call computer vision that's different from human vision. Right? It doesn't have to be you know, 8k, you know what I'm saying? Experience for, an AI application, right? Much lower res. It's an important point that you're making there

Mischa Dohler: 22:40

you can test it out yourself. Just take an image downsize it to whatever resolution and just try different ones and throw it into an LLM and ask him, what do you see? And you'll be surprised. They're pretty good at this. So therefore just getting contextual insights about your, spatial environment with a fairly low frame rate or resolution is okay. Different thing is if you have a, an influence, I dunno, it was like you, actually giving eight K or 4K videos. But even then if you listen to NVIDIA's announcement at cs, they announced that Super re re ls I think it's called, or LSA, I can't remember. you could perfectly four 80 and 78, whatever. And then you have AI basically, upscale that for your viewers to then digest. Yeah. ELSS. Yeah.

Leonard Lee: 23:24

That's upscaling. Yeah, there's a lot of technologies out there that you have to think about as these. Systems evolve. You mentioned Nvidia, you mentioned AI. So inevitably, I have to pick your brain up. What is this guy thinking about when it comes to all this generative AI goodness? I mean, what's your view on it? And, in an overarching way. Not specific to LLMs and things like that.

Mischa Dohler: 23:51

I mean, specifically NVIDIA and Jensen, I mean, the guy is just getting every wave right. I mean, look at CS, he basically went, you know, everybody now is thinking enterprise. He announced a consumer product. I tried to buy it. It was sold out in two milliseconds, you know. And it's basically a super powerful, computer. You can have at home. The only downside is the memory is actually not good enough to run. Let's say the larger models like the deep sea or any of the other stuff, but they will probably address that over time. So he goes from wave to wave and I wouldn't be surprised if the GTC will announce, another big enterprise thing, which will blow everybody out of their mind. Because currently, as we know, other players like AWS and Google can announce their chip fabric. He's somehow always ahead of this. And I'm not surprised because the way how that work is, I think, very market driven, they're not no five year PowerPoint planning exercises. So they're much more nimble to address the market. So I don't work for Nvidia, but I could imagine inside it's a little bit more chaotic than other companies, they get stuff done and, I think they knew already that the training market at some point. We'll go down. It's not only because the cost per token goes down for in appearance and for training, but also we get much more nimble with the training. We have good foundational models. Now is a lot, just fine tuning, but also new models are being announced, which do not require the tokenized approach. So, M. I. T. Stanford, uh, and meta have announced non token based models. So the jury's Still out, which way this sentient AI really will go. And he knows though, that inference will always require compute no matter what. And he will be a winner. Yes or yes. So either it's enterprises needing a lot of compute for inference and training or and, or it will be people doing a lot of inference, no matter what we need, probably GPUs or what type of fabric to really grow. So here you go.

Leonard Lee: 25:46

No, okay. All right. Wow. the guy. And I mean, Jensen, and his jacket, you know, you guys have something in common that's like pretty We do.

Mischa Dohler: 25:54

I took you, it took you a half an hour to realize, was he you first or him? I was first. No doubt. I've been wearing, not this jacket, but you know, I've been going like this for 20 years, so Yeah. So he copied you basically? We did, yeah. Control V straight up. Yeah.

Leonard Lee: 26:10

Well, you know what was weird? I was on an analyst call and I was talk, you know, I, pitched a question to him and he noticed my mac and. I told him, look, I'll give you my Mac if you give me one of your jackets. And so, yeah, I'm still waiting for the jacket. So, hey, you know, NVIDIA AR team, the, the, the offer is still out there.

Mischa Dohler: 26:32

My friend just posted me a mock.

Leonard Lee: 26:36

Well, it looks like he has a new wardrobe based on what he was wearing on. Uh, yeah,

Mischa Dohler: 26:42

that's already not my thing. So now he's really rock star well,, I'm not, I'm maybe in 10 years time. I will, I will change my outfit for the time being. Yeah, I mean, what's your take? Nvidia and, competition is heating up. New chips are being announced. And, yeah, I know every time they announce something, it feels like nobody can catch up with them. And then somebody catches up, go slightly above, and then they do a new announcement. What's your view? You're really in the midst. Oh,

Leonard Lee: 27:05

I think you mentioned deep seek, 1 way or another, they're going to impact the economics of everything and that'll be profound, especially. If what they're claiming in terms of the training costs are irregardless of all the minutiae that people are trying to. you know, get into and determining how they achieved and if they achieved what they did, a bottom line is, there's costs, but then there's price and if the pricing that they put out on the market, just changes the economics, everything. Then I think that's what we have to really look at now, I think is a great opportunity for the folks who want to use, generative AI functions and features and capabilities and embed them into their applications. those are going to be the beneficiaries, but, I think folks still need to be cognizant of the limitation of the technology. It's still very limited, even though we see these benchmarks that show, oh, you know, they're really great. you have to continually consider them with the bigger picture of. Okay, what are you using this for? Does it align with what the technology does? Right? And there is definitely a value that is going to be extracted. Out of this, and I think that's the opportunity you start to look at, because once the economics change, the ROI at the application level becomes different, right? Because a lot of the POCs that we hear in enterprises, yeah, you know, here's some benefit, but it's friggin too expensive. Right? Yeah. Changes. Then we'll see adoption, but it's going to down that upstream. And what I mean, towards Silicon, it's going to impact there. So it could very well be, the event that changes everything and will force all of us to think in a different way about this stuff.

Mischa Dohler: 28:59

Yeah.

Leonard Lee: 29:00

Yeah.

Mischa Dohler: 29:01

It's maybe a good time. You know, I always talk about. Uh, the AA being a real game changer for telecoms, AA standing, for the AAA battery to eject new energy. And one, the first A stands for, ai, the second for API, and the second and third one for ar. So I haven't changed my pitch on this one. I still believe AAA is the way to go for our telco industry. We talked about ar., I haven't talked about API yet, but I strongly believe that we have really managed Ericsson specifically, you know, with the massive investment into Warnoche, even though we had to underwrite quite a bit of this now, but at least people started to believe. So the whole industry now thinks platform for the 1st time, truly for the 1st time. So I'm a big believer that AI really interesting. I would like to highlight maybe the work of. Two people. I really admire. One is my buddy, Demis Hassabis, who is from Google DeepMind. He very quietly, rather than being out there and telling everybody how great they are in terms of AGI for consumers, they're actually doing real things for society, right? He just published the other day, an article saying that their alpha fold has done the research work of 10, 000 PhDs, over three to five years time, they've discovered new materials, which, would have taken us another 500 years to come up with, and they've just done it like this with GPU, but it's incredible. You know, that goes into medicine, real game changes. And that brings me To my another person, I really adore Stanford professor, um, Eric Brimstone, who is, an economist who joined Stanford four or five years ago. I think he will get, Demis got a Nobel prize in, I think it was physics right in the end of the day. I think, Eric will get it in economics, I believe. So he worked a lot on new economic models. And he's arguing very hard is like, if we continue on that, Yeah, the dystopian path of building AI, which is trying to do what we do. We are just undermining the middle class. And then the economic model of modern society will collapse because the value of money has no sense anymore. Right. and he argues we should rather focus AI to do stuff we cannot do. He draws always these three areas, ABC, what humans can do, be a little thing. What we try. AI to replicate, and then C, which is basically the largest blank canvas, where really we should focus AI on and AGI to do things we cannot do. We would never be able to do even with a million years of evolution. So, and that is something I would like us to focus more on and less off some Altman's and deep seek and all that, if it makes sense.

Leonard Lee: 31:39

Agree with you, but that's a tough study. I think it's going to take folks like you and hopefully, me and others to, support that narrative, which I think can, suffer from distraction quite. Significantly, and that's 1 of the reasons why I try to ground the conversation around as much as possible, because if we don't, then we get focused on things that are not necessarily, substantial, and maybe even on the realm of delusional then, we also provide opportunities for the technology to be misused. And I think that's one of the big problems right now okay. So you asked me, what am I expecting out of this year? The other thing is, I'm expecting it for it to be used tremendously in enabling the dark economy. So, being used in cyber attack, cyber crime, and, we're going to see a proliferation of deep fakes. The cyber security industry and the emerging cyber trust industry is going to number 1 be challenged, but, in order for, economies to survive, that trust infrastructure needs to be built in and that's going to, start to become an increasingly important topic for governments. At a government level, because if we can't instill trust in commerce trade and, economy, then all the stuff kind of falls apart. And 1 of the things that always sticks in my mind. And for anyone who goes to any of the Microsoft, events, watch the cyber security segment. This cyber criminal economy, or I think they call it the anti economy or something like that is the fastest growing economy in the world. It is much higher than the U S China, India, any of these mega economies. Wow, bigger and it's 3rd. It went from 8 trillion. I don't know how to calculate this, but, for the sake of discussion, a spouse for a moment, these numbers, 8 trillion to 9. 2 trillion. You do the math that's double digit growth. I think if we give those numbers any credence, that's going to drive, an urgency in reducing the risk and threat that is, associated with that economy. And, to your point, if we can't find the good in AI, people are going to use it to create the alternative economy so that they can feed themselves.

Mischa Dohler: 34:13

Yeah, I know. It's great point. I didn't know these stats. This is really scary, and, um, I was just thinking, you know, our telco industry could really play a strong role because we, we are an industry who actually knows where things happen. So, if somebody was to, let's say, do a deep fake of you with your voice and your etcetera and try to get into your bank account or convince your wife to do certain things, right? So we would know. Where this has happened, whether this is plausible, and in fact, one of the bondage, we acquired this API comes to bondage. One of the APIs is offering this today. So if you combine these 2 things together, we could be a trustworthy. So it's great. Ideally. No, you know what? Oh, my God.

Leonard Lee: 34:50

Oh, my gosh. Oh, my gosh. Yeah. I mean, what a great segue into the API. I think this is absolutely. An important thing, because, you know, I wrote a piece on, the network APIs and what's happening with the Duna, what its role is. Almost 7 years ago, did a study for off calm that you're a chair of, you know, that every decade they do the. The technology futures, right? Well, we want that work, to help off comms research team, figure out, in the five to 10 year horizon from now, what are the technologies and trends that we need to look for? One of the three trends that we talked about was the emergence of the trust infrastructure, right? And so if you think about What the Vonage fraud prevention solution set, which is enabled by network APIs, right? There's also comms APIs there as well, but that actually does dovetails nicely into this idea. But the Duna exchange, I call it an exchange. It provides that, global. Footprint and framework for that trust to happen. And one of the things I cited in my piece was that one of the important things about a Duna is, is going to go through the hard work of, defining and scaling out trust. cause when people think about APIs, they just think, Oh, from a technical perspective, I connect this, this, and then, we can somehow get. frigging 5G connectivity through these APIs are some weird thing, right? But that's not even the half of it. The real reason why you guys have Aduna and you're working with operators to do this is to, is literally, whether or not you guys know it or not, is to create a trust framework so that these APIs can be used globally in a trusted way. But it's interesting that one of those functions, the first use cases, is a trust use case. It's a trust application, you know,

Mischa Dohler: 37:03

yeah, you're spot on. I mean, I'm listening with fascination. I mean, absolutely. Right. And you know, what better industry or what better player to leverage on that trust car than us, right? Because we do have trusted relationships with all the operators around the world. So therefore there is a big narrative there. and then developers will start trusting that as well. And it's interesting, you can't have your application, do one day, not do it, not work one another day. So therefore, I think that's what I do and I will do is, really make sure all these APIs work 24 seven with an SLA of, 99, five nines, whatever, what's. The requirement of the industry. So whoever builds application on top, no, they can really build a proper business case on top, and then make whatever needs to be made. So, yeah, I, you're absolutely right. That trust element never struck me. So clearly until you mentioned that good point.

Leonard Lee: 37:58

Without it, none of this happens and I don't mean just trust in terms of the reliability or trust of the. system, but trust in, do you know who is consuming the, the service, via the APIs and, actually the more, so if you read the piece, obviously you didn't read my piece. I'll send it to you later. Yeah, I will read,

Mischa Dohler: 38:21

of course, I'll, I always read your

Leonard Lee: 38:22

piece when I see them. Yeah. Yeah. It's not just that there's the regulatory bit. When we look at the geopolitical landscape and where things are going, where things are becoming more divided and bifurcated interests are becoming more isolated and alienating, that brokering of trust and, reduction of the friction. Of being able to trust people at is the main function. If you don't have that. doing that, individually, is a tremendous cost. So you kill the entire opportunity that you're looking for. But if you can make that, fabric, a trust fabric, right? For this bigger picture, it's that grease that makes the wheel turn right? That flywheel everyone talks about. And so if you read my piece, that's why I thought this is an essential thing. And it was an interesting. evolution of what you guys are doing with GMP, which I wrote about a long time ago, as well as, the rationale around, buying, Yeah. Yeah. So, surprising was more thoughtful than I thought. Um, it's just that. I don't comment on price, but, the obvious happened. But the thing is that if you look at the intent and the thinking around. Where are you guys and the industry was at the time?

Mischa Dohler: 39:49

Yes, already at the time of purchase. I think a lot of my colleagues, we're really polarized, right? So somehow this ground is of authority as a complete overkill, you know I was one of the folks who had a deep admiration for very actually to do that simply because you know I built companies and I know that sometimes you really need to do a Bold move, to get things moving. And it's very expensive to change the demand side of an industry, whether that's consumer or B2B, when you put 6 billion on the table, the industry will listen. And, it's the price we paid for changing the industry. It's not necessarily the price we paid for getting money back through the APIs. Right. So that platform transformation, which we as a telecom industry have talked about since 4G has now finally happened because of this. So, I thought super bold. I would have done it again and again. Maybe I would have negotiated a little more on the price, and I'm sure the leadership team has done that. It's where we are, but now the industry takes the right way. I'm hopeful.

Leonard Lee: 40:50

Yeah, so, a lot of exciting stuff. We can go on forever. but we're gonna have to cut this off. Otherwise we will go on forever. Uh, this is awesome. we should do this again and again to our audience because, there are some folks who are out there that they want things spoon fed to them. this wasn't for spoon feeding mission. I, yeah, We decided we're going to get together this morning and hit the record button. That's it. hopefully you found some insight and perspectives that were helpful. maybe, enlightening. 1 things for certain, I always have a great time chatting with you. I'm glad we were able to capture a little bit of that lightning in the bottle for our audience here. tell our audience how they should get in touch with you.

Mischa Dohler: 41:40

All right. So, I mean, the easiest thing would be probably, just to drop you a message and then you contact me. Now, I got it back to you. Just, just, yeah, connect on LinkedIn. you know, drop me an email. My email is simple as misha. dollard. com and, happy to discuss further and, you know, if you have any interesting thought, it would be good, to get on a call with Leonard, actually the three of us, the four of us, whatever the five of us, you know, To kind of have a intellectual firepower, a bust on whatever we want to talk about. So it's always great to talk to you later. I just love it. I remember in London, he has a rainy London day. I think last year where we just, earlier with an early morning, but it's nobody that you and me crunching one hardcore topic off another. Uh, it's just such so analog, right? I think I called it an analog morning. It's just beautiful. So that's little than me. So fantastic. Thanks for having me on the show, buddy. Hopefully, you know, we can repeat that one another day.

Leonard Lee: 42:36

Oh, yeah, of course, hopefully it won't take three years hopefully we'll have a lot of folks who you know took a lot from this, Conversation. I know I did it's always great to have a chat with you And by the way, he's a great speaker His daughter is a friggin amazing musician. and just beautiful voice. wonderful artists, such a talented family. And he has the best jackets and the best shoes. I did see your shoes earlier when you pointed your laptop down. So, you know, Yes, you, you delivered on the fashion front. thanks everyone for tuning in, please subscribe to our podcast, which will be featured on the next curve, YouTube channel, check out the audio version on buzz sprout, or find us on your favorite podcast platform. Also subscribe to the next curve research portal at www. next curve. com for. The tech and industry insights that matter. And with that, adios, we'll see you. Thank you so much,

Mischa Dohler: 43:37

Misha. Fantastic. Adios. Hasta la proxima. Thank you, my friend. Cheers. Bye.