Gene Feldman: To tonight's episode of Neon Protocol. This is going to be a little bit different of a typical episode in that it's going to be more of a round table without a guest. So we have our so we have some of our Neon Protocol map committee members here. I never really know what how to refer to us. We're just the members of this wonderful group. so we have Dan Meeks, Mike Van, Damon Shackelford, and Mark Grobe with us. and tonight, rather than having a special guest, we're actually gonna talk about some current headlines that really have been going on that we we've missed. We've been really trying to focus on delivering some really good new content as we're starting this podcast. And so we really wanted to get some really good information out to everybody and start gaining a following and audience, however we want to put it. And we're at a point where we start realizing that there's some really big headlines that we haven't gotten to address. So tonight that's what we're gonna kind of do is talk about some of the recent trends, headlines that have been occurring in XR AI over the past couple of weeks and if not months, and trying to understand what the possible bigger implications could be. So I'm gonna throw I'm gonna kind of start it off with a very big hot take for some of us, meta. And what they've been doing over the past couple months, they went from laying off quite a few of their workforce to move over to AI. Now they're saying they regret it as the announcement from today just came out that they're dropping Ray Band and going to a very price pointed smart glass with their own AI system. So let's talk. What do we implications, thoughts, ideas, opinions? Let's hear it.
Damon: Well, yeah. I'm beginning to hate this company. I really am. I am beginning to despise that leadership. I mean the p there's some really cool people that work there.
Dan Meeks: Yeah. yeah, I haven't had I haven't heard a really good story in like twenty years coming out of Meta, quite to be quite honest. So
Damon: Yeah. There's some cool people who work there, but that leadership sucks.
Gene Feldman: Absolutely. They're changing course right and left and
Dan Meeks: Well we
Mike Vann: So so I've got something and you might have to cut this out, but maybe not. So so how you know, maybe I missed something, I've been busy. Is Kylie Jenner involved with Meta now? I mean, is Kylie Jenner on the payroll? Is that what's happening?
Gene Feldman: that would be a good question.
Mike Vann: Looks like it.
Damon: that's
Dan Meeks: Well well the I I had some some salty quick to give to what is it? An Antonio, the the the blogger that does XR. And he was he was releasing that thing and I and I posed the question, it's like well, Luxotica is a plastic injected company that makes a lot of money and nobody knows why.
Damon: jeez.
Dan Meeks: Anyway. But why why does so much damn money? Like they are just plastic injected glasses, yet yet they they have this, you know, death grip on licensing that they'll sue anybody that tries to copy their stuff and then they charge, you know, five hundred dollars for a pair of glasses. and
Damon: Yeah.
Mike Vann: Well, I mean I think I think the kinda common denominator is fashion, right? Fashion, the way that industry works, the celebrity in and around it and and it's now like all of a sudden upon us in XR. And that's a little foreign, I think, to this industry.
Damon: Mm-hmm. Yep.
Dan Meeks: Hm. Well, I I I
Damon: A lot for and frankly.
Dan Meeks: I guess I picked my wife because she has sensible shoes. I don't know.
Damon: I'm sorry.
Mark G: My hot take my
Dan Meeks: But but no, no, I Mike, you you got a point there, man. I mean, yeah, fashion is a bit of a a fish hook that will grab some people and and really capture them. Yeah.
Mike Vann: It's I I kinda I thought I was being I thought I got got, you know. I thought I was reading an AI story and then I'm like, wait a second, they actually did that.
Gene Feldman: Yeah, and I think it and but you lead up to a good point. It's th think about we've tried everything to to push the technology out. And I think going to Kylie Jenner, from a marketing standpoint, it's actually kind of genius. You know, now for those of us in that have been in industry for a while, you know what,
Damon: Yep, they did it.
Gene Feldman: We had so many cooler things to show. You know, there were so many other things. You know, and just you know, but I do think the price point may actually help. The two nine, you know, as we get past the jokes, the two nine nine price point may actually help them out in the long run. Because I just started reading that because all this news is so new that the LCO or Lexorta actually still make Ray Bans. So they're actually still with the Ray Ban company. But apparently it's so it's just simply a rebranding to a cheaper brand. It's the Aldi glasses. Sorry, Aldi.
Dan Meeks: Wow. Yeah. Really really good European chocolates. Yeah.
Damon: I like Aldi.
Gene Feldman: Yeah, that's what I'm saying. I'm sorry, but it because you well, you
Damon: Yeah.
Gene Feldman: see so many CPG companies selling the exact same product in the aisles of Aldi, and all they do is take off the brand. I mean, like Oreos made by the exact same company. All they do, all they're paying for is the actual Oreo stamp. The only thing it doesn't apply to is Heinz ketchup. There is only Heinz. There is nothing else.
Dan Meeks: Well do you think they're still gonna Do you think they're still gonna push them at the ophthalmologist like they're currently doing? Okay. Yeah, 'cause the the the numbers for the Meta Ray bands, I I really got a feeling that it's because it was an upsell at the the iDoctor that that really got them. So I I wonder if they're gonna be widely available like they were for the Luxotica brands. So that's my question.
Mark G: Well t to your point, that's really the winning aspect to the AI glasses as a whole, right? Even the Roku glasses, right? If my eye doctor has some sort of relationship, you know, I can use my deductible to get the glasses, which is really cool, right? To Meta's point, right, I personally feel it might hurt Meta by getting rid of them Ray Band branding, right?
Gene Feldman: I would
Mark G: Because that's what everybody when I'm wearing AI glasses, everyone thinks it's, are those the Ray bands? Right? And now it's, are those metas? You know, and it's like, okay, so what I'm just gonna quickly say on my hot take is I'm not it's not surprising and it's not unexpected, is that basically, yeah, this is a meta playbook. You know, they buy something, they rebrand it, and then
Mike Vann: Yeah, did you yeah.
Mark G: they'll stop subsidizing it in, you know, four years or two years. yeah.
Damon: Yeah. And and I'll and I'll also support what Mike said at the beginning of this little bit of a tirade, and that is that this is a whole new world for AR, for wearable I mean what are we getting into? Wearable fashion now? We're gonna start seeing runway models with accoutrements you know, to go along with the latest dresses in I don't know, Paris. I don't know. This is very, very new for this market.
Mike Vann: It if it's really gonna go mainstream then this is if this is all gonna come with it, right? And we have to be willing to go for that ride and and then try and help shape where the where the technology can go for everybody.
Mark G: It has to. Yeah.
Dan Meeks: Right.
Gene Feldman: Absolutely. And I do I do think it is going to escalate the the conflict between cameras or no cameras. You know, because right now between the Evan G twos, which congratulations on getting in in the James Bond game, absolutely great marketing play. but you've got this competition now between Evan G two and all the other ones, and it's now to me it's gonna push the question, are the cameras even necessary? And
Damon: Well
Mark G: Mm-hmm. And the going trend right now is the AI glasses being sold in industry or enterprise, all the manufacturers saying it's because there's no camera, right?
Dan Meeks: Well
Gene Feldman: Yes.
Dan Meeks: slight tangent, but I but I also heard some scuttle butt about putting cameras in earpods to do spatial tracking.
Mark G: Yeah, there's this fruit company,
Gene Feldman: I did hear that.
Mark G: right? Fruit company that was gonna do that? Yeah. Yeah, I think there was a fruit company.
Dan Meeks: I'm I'm not sure who who it was. but the idea is that, you know, you're you're essentially you you're cow eyed, you know, you got end up with two cameras on the side of your head and and therefore you're giving more spatial awareness as opposed to something that's locked within your field of view. So that's an interesting take.
Gene Feldman: But my but but my thing is I will never want to watch the videos that I filmed out of my ear canal.
Dan Meeks: That's a good sentence right there.
Gene Feldman: I mean it's yeah, it's it's just it's just not in the cards of any technology dream I've had.
Damon: Edit, edit.
Dan Meeks: Well Well well here's something to think about though. Well it it's more for for the spatial knowing for your AI agent. But but but think of it this way, there there's a lot of YouTubers out there that will use a three sixty camera at a very high resolution and then scale it down to actually frame their video. So it is a way to kind of capture a lot of stuff all at the same time and then make a slightly scale down video that's actually locked to your subject. So that's a possibility. But so yeah, in the future
Gene Feldman: I know.
Dan Meeks: It might not just be the the cameras that are actually on the frames that you gotta worry about. It's where's the camera hidden? yeah, you know, on
Damon: Cameras all around. Yeah. Well, my Nissan right now has has cameras, you know, kind of all around it. And when I hit the reverse gear, it gives me this nice stitched video of what appears to be a top-down view, even though of course there is no camera on the very top of my car. I it really kind of was trippy when the first time it came on. I love it now. It's a great little safety feature. I mean if you go that route with humans, people in a crowd Wanna see who's around you, make sure somebody's not picking your pocket or getting a little too close where they shouldn't be. you know. I could see some spatial awareness being a safety feature too.
Gene Feldman: absolutely. Well, and I mean, even talking about cameras even more, I can't remember which headset that came out off the top of my head, but apparently it took the Vision Pro to the next level by the cameras are actually able to show true eye pass through. So in other words, the blinking you see and everything is not an AI of your eyes. It's actually your eyes coming through the screen coming through the front. I can't remember who it was. It was a big one too. There's been so much movement.
Damon: There's some crazy stuff out there right now. But I guess if Kylie Jenner says it's okay, then yeah, fine, I'll buy it.
Gene Feldman: they're Well, but if but if you can't but it yeah, and I kinda go back to the old marketing back in the early two thousands, if I remember, you know, when we were ch going from pagers to cell phones and there was that sidekick that was like the kind of the in between. It was like the really high tech pager. And I
Mark G: It's very meta of you.
Damon: Thank you.
Dan Meeks: Mm-hmm. And I think they use Parisilton to push that one. Yep.
Gene Feldman: That would be the one. And
Damon: did they?
Gene Feldman: all of
Damon: I forgot.
Gene Feldman: a sudden, pagers and the sidekick blew up. It really did. It it pop culture. It's about bring it which is kinda what we want and we tried everything else, unfortunately. I mean, this is just spaghetti at the wall at this point.
Dan Meeks: Yeah, but I think twenty years on the monoculture's kind of eroded a little bit, at least in the United States, so I don't know if it's gonna stick, you know. But but
Mark G: We hope we hope.
Gene Feldman: Can I
Mike Vann: It'll be quite interesting to see what happens with just unit sales on these things. cameras, no cameras, displays, no displays. I I I have a bunch of pairs of these, as we all do, and I find myself wearing the audio only glasses a lot just because I'm using them as an audio input and they're great. And if I use my AI, you know, apps and agents on there, it's wonderful. when I walk
Damon: Uh-huh.
Mark G: Mm-hmm.
Mike Vann: down to the Panera or, you know, downtown and someone sees cla you know, cameras on my face, I get looks. People are gonna feel that. So it's gonna be really interesting.
Mark G: Yeah. Yeah, that's the aspect, right? It's the risk. It's that risk of, you know, someone sees the camera or they see, you know, the pin light on the one side, you know. Are they gonna punch your block out?
Gene Feldman: absolutely.
Mike Vann: Right? Yep.
Damon: Yeah. It's it's also that nice it's also that nice market data, you know, the finally the user data we've always wanted, right? Instead of pushing the technology, we get enough information to create pull. You know, you don't have to wonder, is this camera too obtrusive? Are these glasses do they look good? will this new hardware venture actually sell? Well, you probably have some data where you could, you know, extrapolate into the market finally. Because you finally have some mass in the market.
Mark G: Yeah, I just going on the whole not surprising, not unexpected, right? Is this just another meta play where they want to totally dominate a market, you know, that's sort of they've created and then, you know, what do you do?
Damon: Mm-hmm. Yeah, meta.
Dan Meeks: Yeah.
Gene Feldman: Ha ha.
Dan Meeks: Small tangent on that. I I listened to a podcast of Alvin Wayne Graylin talking about the history of of the early development of HTC and how meta just completely ruined everything. You know, in back in the day when like the Vibe Pre came out and and all that. yeah, they were they were ahead, but they also didn't have billions and billions of dollars. I think HTC, I think in the podcast you said that the whole company was two billion dollars of HTC. And yeah, meta just completely bulldozed, you know, the next decade. so yeah, Mark Mark's got a point there. You know, maybe that's that's their attempt.
Gene Feldman: Yeah. Absolutely. But the thing I don't like is they bulldoze the they bulldoze the the landscape with all these with what I feel sometimes are very false promises because they're like, all right, we've got this new tool, everybody buys into it, and then a year later they're like, you know what, we're getting rid of it. And it's like how are we ever expected to have anything stick if they're constantly pulling stuff? You know, and
Dan Meeks: Mm-hmm.
Gene Feldman: And to kind of throw back into into kind of where we're going to it with the eyeglasses is I don't think what they really realize the can of worms they also opened when it comes to copyright. So over the past two weeks, there was a great movie that came out. Highly recommend it, if especially if you're Gen X and you grew up with 80s cartoons. It absolutely fantastic movie. But if you go on particular apps that have reels or videos, you will see so many people that just sat there with their glasses and recorded the whole film. The whole film. You know, and you know, then I go back to I'm from St. Louis. We, for those of you that remember in the 90s, Guns N' Roses is not welcome in St. Louis. They have not been for close to 30 years because Axel Rose jumped in the crowd because somebody was taking a photo of him. Which my kids think is so bizarre. They're like, what cameras weren't allowed at concerts? Not at all. And so and so, but I think is what you're gonna have is you're gonna you're gonna see artists, you're going to see movie studios, you're going to see a lot of people. And this is to me, this is a very quiet thing that people aren't I don't think are gonna realize this the type of impact this is gonna have. But but on the flip side, had they had Meta or one of these smart glass companies with cameras giving every Scotsman in Boston or Scottsperson in Boston a pair of glasses, they would be through the roof because we would have so much content out there that would be awesome.
Mike Vann: Yeah. Yeah, there's a lot of change coming. so s smart glasses and everything that's coming in the market there is super hot. What else do we have on the list, Gene?
Gene Feldman: So yeah, we kind of that one kind of went into it. Now, if you guys want to hear more about it too, we did have an episode with Courtney Harding where we really talked about privacy smart glasses. So would highly recommend you guys going back and listening to that. I think the next one is Nvidia really has been making some moves recently. So, Nvidia, you know, because they put Chat GPT and the whole idea of AI out there, now they're actually doing an AI. XRAI public beta. So they're actually now trying to merge the two worlds. Which is what we were always hoping for 'cause we knew they were two sides of the same coin. But now seeing what Nvidia could potentially do with it. So I don't know how much you guys have heard about it. I know it's been relatively recent news. I from what I from what I gather, but I could be way off I could be in a cave.
Dan Meeks: Well, I'm yeah, I'm still trying to rock what you said, both sides of the same coin. What do you mean by that?
Damon: We talked about that and go ahead.
Gene Feldman: So for like for me, I see XR and AI as smaller pieces to a much bigger ecosystem. So in other words, like for me, I it's they're like spokes of a wheel. So we know we've got AI, we've got XR, but to me, there's a higher technology that's above it that is actually comprised of those two pieces in a in a I can't believe I'm gonna say this in a synergetic way.
Dan Meeks: No, I okay. I I think I follow. The the thing about XR is that it it allows for higher levels of abstraction as opposed to like basic computer, you know, programming, all the way up to this, you know, where you're actually just using your hands and, you know, working with the world as as you do as a human being. And so I I'm thinking that's what you're saying, is that it's allowing for that type of computer interaction.
Damon: That was very truthy of you.
Gene Feldman: So it's almost I am Absolutely. It's for me, it's about being able to be Flynn and having Bit sing right next to me, talking to me and guiding me. You know, that I s I see it as two blends that what if, you know, as we create, let's say, these immersive spaces, we're we are so locked into branching. We are so locked into what we can design. But what if we could have AI created as fluid as a dream? Create that experience to where it automatic the AI is automatically able to adapt and create something and cre and adjust the base environment that we create. Holodeck. That's exactly what I'm kind of thinking. Because they would tell the computer, this is the scenario I want. And each time you go into it, it'd be different because the AI is able to have an influence on the immersive experience that helps to create gr greater. believability. Now I'm just talking myself into a circle.
Dan Meeks: Well.
Mike Vann: Well, I mean, I can tell you from the developer side of things, you know, it's great to have options and a lot of large organizations, big enterprises are working with NVIDIA, investing in, you know, equipment and hardware and services. so it's welcome to see them kind of really getting serious about merging XR, digital twins, and AI. because the people want people want to be able to leverage all this super powered hardware that is out there to do great things. And so they seem to be embracing it. there's a tool set more than they had before. so we're actually like a have actually been asked to start to investigate this in some cases. But you gotta you have to have money and and resources to play in that game. But I think what can be done is going to be pretty tremendous.
Mark G: Not
Dan Meeks: You know, clo Cloud XR and Omniverse, all these other prior things, they never really filtered down to me. You know, I don't know if the d direction is down where I'm at, but you know, it was just it's it's just it was never a screwdriver I reached for. You know, I'd I'd fire up Blender instead, you know. So it's it it seems cool, but it's just like, well, you know, may maybe for larger enterprises it's something else, but you know, for
Gene Feldman: Yeah.
Mike Vann: Yeah. Yeah. That's it's kind of the only place it exists and and Dan is on the government and defense side of things, like it may not be there yet at all, right?
Gene Feldman: But the nice thing is it it it to me, it's a movement in a good direction because at least now we can just start gaining content. We can start developing content. Because nothing is worse than having a new device or new software coming and you can't do anything with it. You know, it's like the old video game industry. You know, the c when the consoles would release and there was like the N64, nothing killed it more than Pilot Wing 64. Having that was the only game that you could get for it. You know, it it kind to me it kind of goes back to that. So, but no, getting into that, so leading into what Nvidia's been doing, it seems like this fall is gonna be a big thing with the fruit company we just talked about. So because they have a new s they they're do I mean a huge improvement. I it it kills me took them this long, but I'm not a developer, so I won't say anything. I apologize if I upset anybody on that. But the spatial orb and spatial orb AI and being able to lock things in the Vision Pro OS. So in other words, you can lock different apps to diff to anchor at different places in your physical environment.
Mark G: I just kinda wanna jump in here. I mean, we're kinda you know, I mean just to add a little bit to the multinodal agents thing, I kind of I mean, I'm a little more how do I describe it, right? I'm I'm seeing it from the enterprise side and I really don't see it as a multinodal agents as a big benefit. I really see it as a run on play of service and subscription.
Gene Feldman: No, go for it.
Mark G: For things that some of us in the enterprise space are already doing ourselves. you know, using AI agents in combination to create content, to manage content, to, you know, create a dynamic layer within the experience, I don't think is new. I'm more sort of cautious because, you know, what am I gonna have to pay? Just like with the other AI services, what am I gonna have to pay to get it through? To work? Or am I now going to have to buy a new chipset so that I can run it locally on the VR headset? Right? Call me practical, call me pessimistic, maybe. but that's kind of with the modal nodal agent stuff. In regards to Apple, I mean it's interesting because the whole sort of spatial orb concept, is it really a case of Apple rebranding the concept of spatially aware AI, right?
Gene Feldman: Yeah. And and it's other companies that have done it too, but I think you hit onto a great thing. The trend of subscription-based software has gone overboard. Sorry, this is my hot take. I'm I'm on a soapbox now. because I I've really gotten irate about this because car companies are doing it. As of 2027, car companies are required to have a kill chip in it. That so if the AI senses you are driving erratically, it will kill it. But one company that I won't I I will not say the name. They are four letters, begins with F. But what they have in their kill chip, apparently this this is all alleged this is all allegedly umbert. All right.
Damon: Ha ha.
Mark G: Damn those Italians. Those Italians, I tell you.
Dan Meeks: Ha ha.
Damon: Deep cut. There is the protocol deep cut right there. You got it. Allegedly.
Dan Meeks: Nice.
Gene Feldman: So I'm gonna protect us. Allegedly, allegedly, their their kill chip. I'm gonna have so much fun editing this episode. The kill chip is the kill chip will let you. So if let's say you don't pay, Ford can go in and say, shut the windshield wipers down. Okay, you don't pay for another month, shut the door locks down. You don't pay for a third, they can actually systematically shut down.
Damon: That's gonna be awesome.
Gene Feldman: pieces of your car, but now think about it. Every car you're gonna buy is gonna be fully loaded. And now to Mark's point, you're gonna have to pay a subscription and yearly fees to turn it on and keep it going.
Mike Vann: They're gonna gray out your seat heaters until you upgrade.
Dan Meeks: There there's a key word there. You said buy, but it sure does sound like you didn't really buy it.
Gene Feldman: No. No, you're getting
Damon: That's right. Yeah.
Dan Meeks: Yeah, yeah.
Gene Feldman: it and the thing is w what yeah.
Mark G: The ownership problem, right?
Damon: Uhhuh. Yeah. And the ultimate repo ban.
Mark G: Well no, we'll pick on steam, right? There was a big law case about steam and who owns the game, right? Is it really just a lease? Right? And now the automotive industry is gonna try that, right?
Gene Feldman: Yeah. I mean even to
Damon: Yep. This is this is how you start government regulation and then everybody has fun.
Gene Feldman: Yeah, I mean, and even to the point to where now they're trying to build the vehicles so they cannot be maintained at home. How long before computers are like that? You know, I mean, I mean, we're starting to see just a much bigger thing. You know, and it makes me wonder, okay, at some point you can't get blood from a stone is kind of where I'm going with it. I mean, with these subscription bases. So sorry, I'm done. That's my that's my hot take.
Dan Meeks: Well, you know, it's it's not all doom and gloom, you know, because there there's well, there's there's a term called Liberty Tech, which, you know, there there's all these tinkerers out there. Is it as sexy as all these these big companies? No, but you know, you can get a ESP thirty two board and flash it with some firmware that you found on GitHub and you know, you vibe code an app and you got something that was pretty dang good from from like five years ago. Like right now I got a window up of the mesh tastic, well actually meshcore, people in northern Florida, and we're talking about the weather right now. There's no service, there is no subscription, there's a little antenna on my roof and it's pinging another antenna down the street, which pings another antenna. And we've created our own little network here. And so there there's hope, Mr. Photo. There is some good in this world. You know?
Gene Feldman: No, honestly, I know I I I I'll admit I'm very much
Mark G: Do you believe, Sam? Do you believe?
Gene Feldman: You must believe. All right. Our evangel our evangelistic portion is start it now. No, I I I do ha I do
Dan Meeks: Yeah.
Mark G: Peter Jackson's
Damon: Shun the unbeliever.
Mark G: cut me to hunt us down now. You know that, Dan, right?
Gene Feldman: Yeah.
Dan Meeks: Yeah.
Gene Feldman: No, and I I I believe there's possibility. Like I believe there's gonna be a renaissance where we're gonna see all the the beauty of what this technology can do. But I think right now we're in this point where we're seeing a lot of companies that have have struggled financially and it unfortunately it's it's bearing the brunt on a lot of us. And I think that not trying to get into a political thing, but you know, we're watching companies change so quickly their roadmaps.
Damon: Another infringement.
Gene Feldman: You know, that Google now is going to release glasses. Are they? Are they? You know, I mean that's where I and so I it's we already had so many false starts. Why are we creating more? I think is where I get back to on it.
Mark G: Well, that's because people get desperate when it's not sustainable. That's the real answer. Right?
Gene Feldman: Yeah, but did we not learn from Theranos?
Mark G: No, but that's not that's not what Typro wants. So, you know.
Gene Feldman: Anyways.
Dan Meeks: Yeah, th thus
Damon: Si
Dan Meeks: thus the discussion about Mid Journey and their pivot, you know. I watched a a show, Primogen, by the way, if you ever want to know about what's going on in tech. he he did the math on on the calculations on how the AI would actually function. And it is like many, man many more times than the compute or the network traffic on the internet right now, or it was the other way around, maybe eighteen percent. It was a crazy amount. So, you know, they didn't think it all the way through.
Damon: Yeah, private.
Dan Meeks: Which seems to be a trend, I think. Yeah.
Mark G: Well, you kind of look at like what we've been doing, right? It's the age old story, right? Like enterprise, I'll go circa twenty sixteen or even closer to now, twenty twenty, right? Enterprise was like, can we do it? Right? Then we found value, right? Now it's when I do something in AI, right? I'm being asked, well, how many tokens did it take to do it? Right? Verse
Gene Feldman: Absolutely.
Mark G: Before it was burn baby burn, right? So I think that's really where this all is. It's a case where it's I call it what's the what's the term efficiency first, right? It's now becoming the mantra in enterprise, right? You need to be efficient now. Now because you can do it, now you need to show efficiency, right? You need to show value, real value to usage instead of, I converted a PDF to a Word document and it burned 20 tokens. Right. Now it's a case of no what really did you do? I was sitting on another AI panel And a guy was gloating that he burned a million tokens doing a web page. And I just started chuckling in the presentation. And the speaker said, Mark, why are you laughing? I said, I can't believe you're actually saying you burned a million tokens to make a web page. Right? Because what the value you're showing me, it's actually waste. Actually waste. Because
Gene Feldman: Mm-hmm.
Mark G: A million tokens depending on the framework or model that's doing it, that's a lot of money, right? Compared to you would have got the same results from a firm or studio or marketing group at a quarter of the cost, right? And it goes back to why are they hiring people back? Right? It's because it's efficiency first. So yeah. One hundred percent.
Gene Feldman: and if yeah, efficiency is a big word nowadays. It really is. You know, I mean, watching just a lot what companies do, and this is gonna bring us full circle on the efficiency. Is going back to Nvidia with efficiency. Anybody see the announcement they made about their cooling plant?
Mike Vann: Yeah, completely closed loop.
Gene Feldman: So apparently the Yeah, they apparently they can do a closed loop made of seventy five percent water, twenty-five percent propylene glycol, and it flows through cold plates and it pulls the heat away. It I yeah, if I mispronounce it, sorry, but it takes the cooling related water from two point six million gallons per megawatt each year to near zero in favorable climates. Now, I question the flavor in favorable climates. I'm in Missouri.
Mark G: Problink like
Gene Feldman: There are no favorable climates here. Okay. Period.
Damon: Ha ha ha
Gene Feldman: So let's just start there. But that if they could I mean, think about the resolution to a huge argument going on right now, which is the data center cooling. Yeah, and I don't know if it's as big as a issue with where you guys live. It's because I'm in Missouri. I mean, we have a town Festus that's I mean making national news 'cause they're just going head to head with like with a major data center provider.
Mark G: Yeah, North Jersey towns are towns are vetoing left and right. Yeah. You know, I kinda joke that the the think of the old cell tower, right? The old cell tower problem. You know, carriers wanted to bring cell towers into small towns and they would basically just destroy it because they're like the beauty of whatever, right? Same thing here.
Mike Vann: It's happening everywhere. Yep. But very nearby.
Damon: Yeah.
Gene Feldman: absolutely. Well, and the or the wind towers. But the one thing I keep going back to is I just blinked out of it too.
Mike Vann: I mean, while you're thinking, Gene, it's an interesting point. I mean, these are huge challenges that, you know, a lot of really massive companies are going after now. And they will solve them. They'll come up with a lot of novel ways to do it. So part of the benefit here is that some technologies are going to be developed. That heat transfer thing, I mean, they've already been doing this for years, but they're going to do it at such a level that, you know, that can start to maybe impact the other energy needs and and energy production for the rest of the world.
Gene Feldman: Yeah, go for it.
Mike Vann: And you know, some good is gonna come out of solving some of these challenges. So I think there's some good you know, side product of all of this. I mean there's probably as much or more bad, but hopefully, you know, if you look at it optimistically, we're gonna invent some new technology along the way.
Gene Feldman: absolutely. 'Cause I just look at this as a as a of evolution of bitcoin bit Bitcoin mining.
Mike Vann: Yeah. Yeah.
Gene Feldman: Yeah.
Dan Meeks: To to Mark's point, you know, it's all
Gene Feldman: Yeah.
Dan Meeks: about efficiency, right? So this this build build giant plants and and make everybody angry and burn a lot of electricity and at least the water seems to be solved. But there's gonna be this march to keep making this more and more efficient because right now this isn't gonna work. so I I bet top men are working on this at the moment.
Damon: Sure. Topman, another deep protocol cut. That was not in the movie. Sti stick with no stick with the deep cut.
Gene Feldman: Now to be fair, top men and women are working on this. Let's let's let's be fair. I I had to give a twenty twenty six edit to it. We're only seven episodes in. Let's be good.
Damon: Ha ha ha.
Dan Meeks: Now now now it is it it would be recursive if if the AI is trying to figure out how to make better AI. That would be kinda cool. But who knows? Maybe they are. Yeah. It's good for that kind of thing.
Damon: That's actually already happening too. Yeah, that's anthropic's
Gene Feldman: Yeah.
Damon: problem right now.
Gene Feldman: Well well, even the a company we talked about previously, they had that same problem a couple of years back where they actually had to shut the AIs off because they developed their own language. You know, to talk to each other. And like I always well, I tell people, if you really want to have a really fun experience, take Siri, have her ask Alexa a question. As Alexa responds, what'll happen is within about 30 seconds, they will go from talking in English. And once they realize they're talking to each other, they go into into sounds into basically binary speak of tones. Is the coolest thing in the world, but it really scares you.
Mark G: Will my will will my AI glasses
Dan Meeks: I saw that episode of Star Trek. Yeah.
Mark G: translate that discussion? Mm.
Gene Feldman: that's a g that's a fourth wall break right there.
Mike Vann: That's a product. They go after that one, Mark.
Gene Feldman: All right. So if you try it, 'cause if you if you're if you're listening and you try it, please send us the response. We would love to hear what happens.
Mike Vann: That's a movie.
Damon: Tossed Mark. Assuming you're still alive when it finishes.
Gene Feldman: Yeah, it's kind of like ty typing Google into Google. You will break the internet.
Dan Meeks: Divide by zero. Yeah.
Gene Feldman: So all right. So let's see. So no, I think I mean I mean honestly, I think we've really kind of talked about, you know, some of those those big things, but it it kind of goes, I think, to the bigger to the overall theme of just how fast things are moving again. And how, you know, I mean I mean, I remember a year ago we were all talking about
Damon: what's next? Do we want to beat that a little more?
Gene Feldman: Before we actually start the podcast and then just start normal conversations, we had this grand idea of what do we call it? You know, what's this new what's this what's a better word? And seeing spatial, which honestly I didn't think I'll admit I was I was one of those people that didn't see I thought it was more of a work relate than kind of this overall computing, but really seeing it resonate with people. as a idea, I think is is really cool that and I think for me it's it's seeing things that are moving so fast, but also understanding that you know it it's growing organically by people now developing their own language for it. Does that make sense? Or did I just get really deep and philosophical on a Wednesday night?
Mike Vann: I think people I think people with all levels of kind of grasp on this stuff are coming up with their own ways to explain it, right? And that's all the way down to like person on the street that has nothing to do with technology. So I think that's kinda natural. this stuff is going mainstream, everybody's using it in some way. so it's pretty interesting to watch. you know, there's a tremendous amount going on and that's gonna happen and we're in the middle of it, which is
Dan Meeks: Up. Yeah. Yeah.
Mike Vann: fun time to be in tech and and yeah. I mean I'm glad we're doing this.
Dan Meeks: Yeah, the the term glaze. You know, I used to talk about donuts. but but now, you know, that's a term of of the AI being just a little too nice to you. you know, that that entered the lexicon. I don't know where that that bubbled up from, but it was definitely organic, so
Gene Feldman: Yeah, yeah, yeah.
Mark G: Whoa. So you're saying to me now glazed donut is how we now describe false confidence of AI? Is that what you're saying? Just glaze. Glaze.
Damon: I think just glaze. No, just the word glaze.
Dan Meeks: Y Yes. I don't know. J just glaze, yeah. Yeah.
Damon: The ans that that answer was glazed.
Mark G: I learned something today.
Gene Feldman: You know what? That's on fleek.
Damon: Six seven.
Dan Meeks: Yeah. Well and also the derogatory term clanker, you know, I say it with a hard R, to to my wife, you know. She'll she'll she'll say like this damn clanker just glazed me again. It's like we did not say that to each other two years ago. So so but that's where we are. And so to your point, Gene, I mean you're we don't know what the future is and we're we're kind of stumbling into it every single day.
Damon: But no, Mike's right. This is a go ahead.
Dan Meeks: And and to kind of bring it to a point, like, I I had a real heart to heart with Claude the other day, because I was like, Hey, I ain't hanging my ID over to you to keep working with you. even though you know everything about my business and have taken pictures of my front yard because I work on computer vision, like you have all this stuff. Why do you need me to humiliate myself with an ID? and it tried to calm me down from the ledge saying, No, no, no, you're you're not an Iranian terrorist, Dan. It's it's cool.
Damon: Yeah.
Dan Meeks: that's that's really what it's about. And I'm like, yeah, but still. But anyway, I don't know what's gonna happen next. So I'm thinking to myself, like, okay, do I need to take some money out of savings and get that local AI agent spun up sooner than later? Because every my my entire career of technology is all about rug poles. And I'm like, Well, is the rug pull happening? Like is is this the beginning of it? Because I sure am invested like like Claude's been a huge help with me, you know, just being a single person business.
Damon: Yeah.
Dan Meeks: Wow, it's been a force multiplier. but it might go away at any moment. So I gotta be prepared.
Mark G: Yeah, yep, it's time Yep, it's time to dust off time to dust off Deep Seek. Run it local. Yep.
Gene Feldman: Yeah.
Dan Meeks: Mehoo. Yeah. So yeah.
Mark G: Alibaba will thank you.
Damon: Never I'm never going back.
Gene Feldman: Yeah. But but but you bring I mean, but you bring up a great point. I mean, 'cause as we've seen with other companies, what's to stop Nvidia from going to Claude and saying, you know what? We're gonna give you XY XX amount of money. And you know what? If the money's good enough, they'll sell it. And next thing you know, Claude's either integrated or shelved. And you know, so it's it's really especially what's gonna happen at some point are and I always always like with VR too, is the ecosystem. Are we ever going to have a point where the AIs can talk to each other? That, you know, we already hate shifting our cell phones, especially if we have to shift from Apple to Samsung. Now, like you, you have all your stuff in Clawed. If you had to go to a different AI engine tomorrow, the amount of work is unfathomable. Yeah.
Dan Meeks: Yeah, that's gonna be a lot of markdown files to to build
Damon: No. Nah,
Dan Meeks: the context again. Yeah. So
Damon: there's yeah, there's a there's a command now. You can you can download your history and switch files over. If I want to switch I switch between Grok and Claude regularly.
Gene Feldman: But I just I just I guess I just look at as it would be great if they they could talk to each other where you didn't have to do that. If that makes sense. Where, you know, it basically it's an AI ecosystem. But does that require uniformity? So in other words, one AI does it th will there ever be one ring to w rule them all?
Mark G: Well, that's the original hype, right? That was the original hype was, you know, that each chat GPT was the best AI or Claude was the best AI, right? You know, now you're starting to talk about with the developing concepts of trust layer, right? And the aspect of, you know, now we need specifically in enterprise, right? We need some sort of layer to be created where one, we can manage the trust, but also manage the ability to do flip flop based on
Damon: Not with this much money involved.
Dan Meeks: Do you want that?
Mark G: what we're doing, right? So I call it the use case layer, right? And this is kind of like starting, right? Because now if you're making me pay to use this product, right, now I'm going to be super critical. Right. If I'm paying Cloud or I'm paying Copilot or I'm paying WatsonX or something else, right? You know, I want to make sure I have my interoperability because I understand there's a benefit. To each of the models. So I should be able to capitalize on. Right? Because now I'm paying. So to your point, right? That's an additional layer.
Gene Feldman: Yeah, and and you know what? You for the for the benefit of that you have to pay a subscription. And we've come full circle again.
Mike Vann: Well I I could you know, these things are being worked on and they are not coming from those five trillion dollar companies. They're there are people developing these kind of independent layered solutions and you know, whether or not they can emerge and be significant, maybe maybe for business use and for for enterprise in ways where it gives people and companies kind of their own control. that those could become real products, but you know, it's all about trust and it's all about risk and legal and all the things that companies care about, right, Mark? So yeah, I mean I think there's gonna be a lot of that, but you you're gonna have Google and Microsoft and Anthropic and everybody continuing to bang down bang your doors down and demand your money. And you know.
Mark G: But here here here
Dan Meeks: Yeah.
Mark G: here's here's the flip. Here's the flip.
Gene Feldman: absolutely.
Damon: Yeah. And and now China's coming out with some really interesting stuff too. So add them to the list.
Mark G: So so what happens when, you know, we were just talking about, you know, we take we take the technology in a full circle, right? So now what happens when, as a business owner, right, NVIDIA has rights to your solution because we're talking about small guys figuring out the right way, right? What happens when NVIDIA says, Well, we're gonna take away your tensor for that solution? Right, because there's a dependence of platform, right? What happens? Right? That's like the old AWS problem with small business, right? There were small businesses that really created an awesome solution. So what happened? Amazon came over and said, we're gonna charge you more, you know, there's more service fees or whatever, to the point where they took it, right? Because it worked. So the question now is it when more so than if, when is that gonna happen with something like AI, right? And What's gonna ha what's gonna become of it? Is that gonna be the pop? Is that gonna be the bubble? Right? Trigger or is it something else?
Gene Feldman: absolutely. Plus, I keep thinking about that we're not even anticipating what these kids that are learning AI naturally, you know, what I mean, in five years from now, when these 12 year olds get through high school and, you know, and all of a sudden they enter, what type of things are they gonna be creating homegrown that are gonna be ten times better than anything? You know, and
Dan Meeks: Well y yes and no, because yeah, they're gonna be they're gonna be totally adapted to this new world, but with zero knowledge of the fundamentals. And I tell you, man, I I I saw it in the Air Force. So what we did, and make sure I don't get on this for too long, but I was right at the end when we did celestial navigation in aircraft, where you put a sextant up on the top of the aircraft and actually shot the the the sun and the moon and Arc Touris and all that stuff.
Gene Feldman: Good point. No, please.
Dan Meeks: I was at the tail end of that. And when I came back as an instructor, we were we were now combat systems officers, meaning you were in charge of the GPS and the inertial navigation system and all these other things that you had on board. But you only had instead of three months about the fundamentals, you got two weeks. And I saw that in my students that they didn't really understand what was going on underneath the the hood. So I don't I don't have an answer though. Like like Yeah, they're gonna be really awesome at at using AI, but they're also not going to have something because there's only so many penguins you can put on your iceberg of a brain. So, you know, one jumps on, another one's gotta jump off. So you know, that's how it goes.
Damon: Ha ha ha.
Gene Feldman: no, I agree. And I think that's where like the experience I think as kind of getting back to a previous conversation, like as we throw immersion into it now, that can help give them that real time experience of you know, to to help kind of offset it. Because I do think it's, you know, like from a learning standpoint, this could be an equalizer. And I only say it because there was a recent use case I came across with some firefighters that I I found Really interesting. I actually shared it on LinkedIn. But it was a study out of Brazil that where they said experience barely touched performance when using this new technology for VR, like VR training with AI. They were saying that what actually moved the needle, you know, so they all kind of came to even the 20-year veterans of really not under knowing how to use the technology. You know, the VR headsets, the AI, they were all kind of coming to it was. Because firefighting very experience-based type of role. But what they found is that in 120, they tested 126 firefighters in VR. The years of experience that they had explained less than 1% of their full performance along with their gaming experience. So basically, They didn't need experience with the not with the with the VR headsets or the type of gaming that was done. Once they got in there, they about 98% of them said the realism was so powerful from what was being created and what was being shown that it surp they immediately jumped into firefighter mode. And that because the stakes were so high or the believability, and I thought that was a really interesting use case. Because what it really did is show that you could have all the experience, but if it's the first situation you're ever experiencing, regardless of twenty, thirty years, you're still coming at it new. And they were able to approach that. So you have, let's say, a new firefighting method. Everybody's got to learn it. What a great equalizer across the learning models.
Mark G: mean I I mean I I I think what you're describing to me is what I traditionally would call just the accelerant, right? It has nothing to do with AI. It's the fact that the simulation that you're using for training was properly designed. Hence you have the accelerant, right? You know, the was a I I'm trying to think of the military term, Dan, help me out. You know, it was a force multiplier. There we go. Right. you know, I think that's really what
Gene Feldman: Sorry, I know I got really long winded on that.
Dan Meeks: I don't know. No, that's that's interesting, yeah.
Damon: Ha ha ha.
Gene Feldman: Yeah.
Dan Meeks: Mm-hmm. Yeah.
Mark G: you're talking about. It really has nothing to do with AI, right? I wish we had Bob Stone on here, right? 'Cause Bob would be the guy that'd actually probably just say what I said, right?
Gene Feldman: Yeah.
Dan Meeks: Well, if if I can give a a a quick war story to to kind of amplify what you're saying there, Mark, is so part of my work, because w you know, the Afghanistan war dragged on for a long time, there was a village just south of Kabul that did most of the JTAC training in the simulators. And so I spent a good decade in that village building missions, testing stuff out, training people, And then I got on a on a treadmill. It was an omnifinity. And my brain clicked. Like there was a whole nother side of my brain that activated. Because I had been walking down those alleys and streets for a decade. But it wasn't until I actually put one leg in front of the other that the depth of the alley f clicked. Like all these other things, you know, I was in a headset, obviously, but you know, it It wasn't until my body was fully engaged that the rest of my brain, you know, saw what was happening. So yeah. Designing the simulation correctly for for what you need to do.
Gene Feldman: absolutely. But and I think this is where and this I think I can exp I know I'm coming a broken record on it, but you know, of how the AI piece can come into it is by creating additional functionality of things to so people don't have to create. So a great example. And that Dan, this is I'm right in there with you. When you have something that clicks and all of a sudden it's like, all right, I'm there. For me, I had an opposite reaction to one one particular app. I loved it. It was an old on the Oculus when it first came out the rift. It was the Blade Runner apartment. I was so excited to go into it and it was beautifully done. Who if the artists happen to ever listen to this podcast, you guys really did a great job. However, when I went to the piano and I couldn't play anything, when I went to the refrigerator and I couldn't open the door, all of a sudden that realism shattered in a heartbeat. And so for me, it also goes on the side effect. And that's where I think AI can come into it is by saying, all right, you know what? The user wants to open this. We never anticipated anybody ever wanting to do it, but let's go ahead and let them do it. You know, it's same thing like the open world games. And so that's where I kind of come into it because we also never want the illusion shattered as well. You know, for a deep cut, we never want somewhere in time penny moment to. ever happen to any of us. For those of you who never saw the movie Somewhere in Time, watch it and you'll understand the penny moment.
Mark G: So but what you're describing to me is what we do now in enterprise, right? The whole point is the AI sort of automates or provides a sort of dynamic response to scale for the way a user say interacts unexpectedly, right? So I see the value that you're talking about, but to me it's just scale.
Gene Feldman: Yeah. Absolutely.
Mark G: Right. It's just it's just a function of scale.
Gene Feldman: Yeah. really?
Dan Meeks: Yeah, but Mark, you you
Gene Feldman: Absolutely.
Dan Meeks: you've raised this question before, which is it's dynamic, but is it consistent? Because, you know, when when you're training somebody, you know, and I I think we we had a prior discussion on this before, like in the Air Force, you you gotta give everybody a fair shake because if you wash them out, you gotta say, Well, we we did everything that we're supposed to do and you weren't up to standards. But if you say, Well, you know, we we went above and beyond with this student. And he barely got his wings, but this guy, you know, we kind of half ha half assed it and you know, we we washed him out anyway. You you can't you can't do that. So that's the problem with the AI though, is getting consistency every single time. And if you're in an enterprise side where you're trying to be consistent with your metrics, how do you how do you make that happen? I don't know.
Mark G: Mm-hmm.
Gene Feldman: Yeah. And then and to throw a bigger wrench into it, for how do you also account for those one off situations where you have to be inconsistent with how it's been done? You know, you know, like for me, it's like I work on motorcycles. Ninety nine percent of the time this will work. But I you know what but I work on again, motorcycles. It's always that one percent time it doesn't work that we've got to figure it out.
Dan Meeks: well yeah. Well that's why in the Air Force we you we we have human beings doing the instructing because they're they're stuck within a a framework and and what they're really allowed to do, but they they know where the walls are. so but in AI you're you're never really quite sure about that. So
Gene Feldman: Yeah.
Mike Vann: It might
Mark G: Yeah, I mean you have thresholds
Gene Feldman: Absolutely.
Mike Vann: I mean I think
Mark G: of bias, but that's about it, right? You know.
Dan Meeks: Yeah, me and Walls. Yeah.
Gene Feldman: Yeah.
Mike Vann: I mean this is a great topic. Just just as a point for Neon Protocol, let's let's look at getting some people in that are really doing th innovative things with the kind of convergence of AI and XR. We're we all have done some of this. there are people out there doing cool things. I don't it's not appropriate for every industry, every business, every use case, but there are some interesting things being done. It might be interesting on one of the episodes to aggregate some people that are out there doing those things.
Gene Feldman: Absolutely.
Damon: Mm-hmm.
Gene Feldman: No, in fact,
Mike Vann: And I'm sure they would love to come on.
Damon: Agreed.
Gene Feldman: yeah, absolutely. In fact, that that's a great plug for our next episode. Our next episode's gonna feature Sophie Thompson from virtual speech, which this is what she's actually this is her bread and butter, which is designing so her her story is really interesting. I'll let her tell most of it, but kind of give you a background. So when she was in university in England, she hates public speaking. She absolutely despises it.
Damon: Yeah, yeah.
Gene Feldman: But so she wanted to be able to practice. So she created an AI bot, an like a chat bot with a face and everything. That way she can practice giving speeches and realize that it could grow more from that. So sh straight out of college, she started virtual speech. And what she's doing is using AI avatars to create interactive simulations that respond in real time. based on whatever the AI has been without whatever data the AI has been loaded with. So and you can set the temperament of the simulation. So where I found it be really impactful, she's got a great simulation that's an elevator pitch. You've got 60 seconds on an elevator ride to pitch a CEO. That's it. And I mean you're in the elevator. It's going, I mean, it's beautifully done, but it's a great blend of this conversation. of what we're talking about, and I I know she's had some really great results that I know I think she's really excited to share with us.
Mike Vann: Good.
Dan Meeks: Now if I do that in XR and I
Damon: Cool. Very cool.
Dan Meeks: hit the stop button and the elevator doesn't stop, I've lost immersion. Just kidding. No, no, it sounds cool though. That sounds really cool.
Mike Vann: You shoot through the top of the building.
Gene Feldman: No, my my yeah, my luck is they're only going one floor. It's like, all right, yeah. So you're you're following behind, you know, with your going, please, please listen. No, so no, so yeah, I agree. And in fact, one of the things I wanted to do too as we're kind of wrapping up is first call out, we've had two people that have actually listened to every episode so far, Dorian and Kelly. So if you guys are listening, thank you guys for the support. We're really appreciating it. Kelly's actually gonna be on an episode in a couple months. He actually went to high school with me, so we're gonna try to so not to surprise anything, but we're gonna try to have an episode talking about technology through the ages as we saw it develop throughout our our experiences. So but I would love to also have him on because he talks about AI a lot and works in it. But If you're out there and you're listening and you have a really good use case that you want to share or you think that would be worthwhile to add to the Neon protocol, please send us a message. we would love to talk to you and potentially include your use case or your opinions. As you see, we kind of have some strong ones. but your opinions, your thoughts, even if you know, what even your results, where you're going, because again, we can only learn from each other. So,
Mark G: So right now, subscribe and drop it in the comments.
Gene Feldman: Absolutely.
Damon: Yeah, we're ha yeah. We need to give a way to get in touch with us.
Dan Meeks: Hit that notification bell.
Gene Feldman: Absolutely. So well if you want to get in touch, what you can do is leave go to Alachkey Nation on LinkedIn and you can actually add a post or you can actually start a communication.
Dan Meeks: Yeah.
Mark G: Comment comment in the video comment in the video
Damon: Marcus, comment comment comment
Gene Feldman: I I will cut it yeah.
Damon: in the video? Comment in the video. Do we allow comments in our videos? Subscribe and comment. Like and share.
Gene Feldman: All right.
Mike Vann: Subscribe and comment.
Mark G: Subscribe and comment.
Gene Feldman: I I I will cut all that out. Yeah. So
Mark G: Yeah, like and share. That's it.
Gene Feldman: Yeah, so if you want to add, so please comment, like, share, follow, subscribe. Please, we're really trying to get this to grow. And we've got seven episodes out, and so far we've had a really even though it's a one hour podcast on tech, we've had really good responses and we really wanna make sure that we're we are covering everything as beyond just our opinions. So with that, we thank you guys for joining our episode tonight. if you want to learn more, please again go to link go to our LinkedIn page. Feel free to start a conversation. We would love to talk to you guys. And in the meantime, we'll talk to everybody soon and enjoy technology. It's about the best I can say.
Damon: Well, yeah. I'm beginning to hate this company. I really am. I am beginning to despise that leadership. I mean the p there's some really cool people that work there.
Dan Meeks: Yeah. yeah, I haven't had I haven't heard a really good story in like twenty years coming out of Meta, quite to be quite honest. So
Damon: Yeah. There's some cool people who work there, but that leadership sucks.
Gene Feldman: Absolutely. They're changing course right and left and
Dan Meeks: Well we
Mike Vann: So so I've got something and you might have to cut this out, but maybe not. So so how you know, maybe I missed something, I've been busy. Is Kylie Jenner involved with Meta now? I mean, is Kylie Jenner on the payroll? Is that what's happening?
Gene Feldman: that would be a good question.
Mike Vann: Looks like it.
Damon: that's
Dan Meeks: Well well the I I had some some salty quick to give to what is it? An Antonio, the the the blogger that does XR. And he was he was releasing that thing and I and I posed the question, it's like well, Luxotica is a plastic injected company that makes a lot of money and nobody knows why.
Damon: jeez.
Dan Meeks: Anyway. But why why does so much damn money? Like they are just plastic injected glasses, yet yet they they have this, you know, death grip on licensing that they'll sue anybody that tries to copy their stuff and then they charge, you know, five hundred dollars for a pair of glasses. and
Damon: Yeah.
Mike Vann: Well, I mean I think I think the kinda common denominator is fashion, right? Fashion, the way that industry works, the celebrity in and around it and and it's now like all of a sudden upon us in XR. And that's a little foreign, I think, to this industry.
Damon: Mm-hmm. Yep.
Dan Meeks: Hm. Well, I I I
Damon: A lot for and frankly.
Dan Meeks: I guess I picked my wife because she has sensible shoes. I don't know.
Damon: I'm sorry.
Mark G: My hot take my
Dan Meeks: But but no, no, I Mike, you you got a point there, man. I mean, yeah, fashion is a bit of a a fish hook that will grab some people and and really capture them. Yeah.
Mike Vann: It's I I kinda I thought I was being I thought I got got, you know. I thought I was reading an AI story and then I'm like, wait a second, they actually did that.
Gene Feldman: Yeah, and I think it and but you lead up to a good point. It's th think about we've tried everything to to push the technology out. And I think going to Kylie Jenner, from a marketing standpoint, it's actually kind of genius. You know, now for those of us in that have been in industry for a while, you know what,
Damon: Yep, they did it.
Gene Feldman: We had so many cooler things to show. You know, there were so many other things. You know, and just you know, but I do think the price point may actually help. The two nine, you know, as we get past the jokes, the two nine nine price point may actually help them out in the long run. Because I just started reading that because all this news is so new that the LCO or Lexorta actually still make Ray Bans. So they're actually still with the Ray Ban company. But apparently it's so it's just simply a rebranding to a cheaper brand. It's the Aldi glasses. Sorry, Aldi.
Dan Meeks: Wow. Yeah. Really really good European chocolates. Yeah.
Damon: I like Aldi.
Gene Feldman: Yeah, that's what I'm saying. I'm sorry, but it because you well, you
Damon: Yeah.
Gene Feldman: see so many CPG companies selling the exact same product in the aisles of Aldi, and all they do is take off the brand. I mean, like Oreos made by the exact same company. All they do, all they're paying for is the actual Oreo stamp. The only thing it doesn't apply to is Heinz ketchup. There is only Heinz. There is nothing else.
Dan Meeks: Well do you think they're still gonna Do you think they're still gonna push them at the ophthalmologist like they're currently doing? Okay. Yeah, 'cause the the the numbers for the Meta Ray bands, I I really got a feeling that it's because it was an upsell at the the iDoctor that that really got them. So I I wonder if they're gonna be widely available like they were for the Luxotica brands. So that's my question.
Mark G: Well t to your point, that's really the winning aspect to the AI glasses as a whole, right? Even the Roku glasses, right? If my eye doctor has some sort of relationship, you know, I can use my deductible to get the glasses, which is really cool, right? To Meta's point, right, I personally feel it might hurt Meta by getting rid of them Ray Band branding, right?
Gene Feldman: I would
Mark G: Because that's what everybody when I'm wearing AI glasses, everyone thinks it's, are those the Ray bands? Right? And now it's, are those metas? You know, and it's like, okay, so what I'm just gonna quickly say on my hot take is I'm not it's not surprising and it's not unexpected, is that basically, yeah, this is a meta playbook. You know, they buy something, they rebrand it, and then
Mike Vann: Yeah, did you yeah.
Mark G: they'll stop subsidizing it in, you know, four years or two years. yeah.
Damon: Yeah. And and I'll and I'll also support what Mike said at the beginning of this little bit of a tirade, and that is that this is a whole new world for AR, for wearable I mean what are we getting into? Wearable fashion now? We're gonna start seeing runway models with accoutrements you know, to go along with the latest dresses in I don't know, Paris. I don't know. This is very, very new for this market.
Mike Vann: It if it's really gonna go mainstream then this is if this is all gonna come with it, right? And we have to be willing to go for that ride and and then try and help shape where the where the technology can go for everybody.
Mark G: It has to. Yeah.
Dan Meeks: Right.
Gene Feldman: Absolutely. And I do I do think it is going to escalate the the conflict between cameras or no cameras. You know, because right now between the Evan G twos, which congratulations on getting in in the James Bond game, absolutely great marketing play. but you've got this competition now between Evan G two and all the other ones, and it's now to me it's gonna push the question, are the cameras even necessary? And
Damon: Well
Mark G: Mm-hmm. And the going trend right now is the AI glasses being sold in industry or enterprise, all the manufacturers saying it's because there's no camera, right?
Dan Meeks: Well
Gene Feldman: Yes.
Dan Meeks: slight tangent, but I but I also heard some scuttle butt about putting cameras in earpods to do spatial tracking.
Mark G: Yeah, there's this fruit company,
Gene Feldman: I did hear that.
Mark G: right? Fruit company that was gonna do that? Yeah. Yeah, I think there was a fruit company.
Dan Meeks: I'm I'm not sure who who it was. but the idea is that, you know, you're you're essentially you you're cow eyed, you know, you got end up with two cameras on the side of your head and and therefore you're giving more spatial awareness as opposed to something that's locked within your field of view. So that's an interesting take.
Gene Feldman: But my but but my thing is I will never want to watch the videos that I filmed out of my ear canal.
Dan Meeks: That's a good sentence right there.
Gene Feldman: I mean it's yeah, it's it's just it's just not in the cards of any technology dream I've had.
Damon: Edit, edit.
Dan Meeks: Well Well well here's something to think about though. Well it it's more for for the spatial knowing for your AI agent. But but but think of it this way, there there's a lot of YouTubers out there that will use a three sixty camera at a very high resolution and then scale it down to actually frame their video. So it is a way to kind of capture a lot of stuff all at the same time and then make a slightly scale down video that's actually locked to your subject. So that's a possibility. But so yeah, in the future
Gene Feldman: I know.
Dan Meeks: It might not just be the the cameras that are actually on the frames that you gotta worry about. It's where's the camera hidden? yeah, you know, on
Damon: Cameras all around. Yeah. Well, my Nissan right now has has cameras, you know, kind of all around it. And when I hit the reverse gear, it gives me this nice stitched video of what appears to be a top-down view, even though of course there is no camera on the very top of my car. I it really kind of was trippy when the first time it came on. I love it now. It's a great little safety feature. I mean if you go that route with humans, people in a crowd Wanna see who's around you, make sure somebody's not picking your pocket or getting a little too close where they shouldn't be. you know. I could see some spatial awareness being a safety feature too.
Gene Feldman: absolutely. Well, and I mean, even talking about cameras even more, I can't remember which headset that came out off the top of my head, but apparently it took the Vision Pro to the next level by the cameras are actually able to show true eye pass through. So in other words, the blinking you see and everything is not an AI of your eyes. It's actually your eyes coming through the screen coming through the front. I can't remember who it was. It was a big one too. There's been so much movement.
Damon: There's some crazy stuff out there right now. But I guess if Kylie Jenner says it's okay, then yeah, fine, I'll buy it.
Gene Feldman: they're Well, but if but if you can't but it yeah, and I kinda go back to the old marketing back in the early two thousands, if I remember, you know, when we were ch going from pagers to cell phones and there was that sidekick that was like the kind of the in between. It was like the really high tech pager. And I
Mark G: It's very meta of you.
Damon: Thank you.
Dan Meeks: Mm-hmm. And I think they use Parisilton to push that one. Yep.
Gene Feldman: That would be the one. And
Damon: did they?
Gene Feldman: all of
Damon: I forgot.
Gene Feldman: a sudden, pagers and the sidekick blew up. It really did. It it pop culture. It's about bring it which is kinda what we want and we tried everything else, unfortunately. I mean, this is just spaghetti at the wall at this point.
Dan Meeks: Yeah, but I think twenty years on the monoculture's kind of eroded a little bit, at least in the United States, so I don't know if it's gonna stick, you know. But but
Mark G: We hope we hope.
Gene Feldman: Can I
Mike Vann: It'll be quite interesting to see what happens with just unit sales on these things. cameras, no cameras, displays, no displays. I I I have a bunch of pairs of these, as we all do, and I find myself wearing the audio only glasses a lot just because I'm using them as an audio input and they're great. And if I use my AI, you know, apps and agents on there, it's wonderful. when I walk
Damon: Uh-huh.
Mark G: Mm-hmm.
Mike Vann: down to the Panera or, you know, downtown and someone sees cla you know, cameras on my face, I get looks. People are gonna feel that. So it's gonna be really interesting.
Mark G: Yeah. Yeah, that's the aspect, right? It's the risk. It's that risk of, you know, someone sees the camera or they see, you know, the pin light on the one side, you know. Are they gonna punch your block out?
Gene Feldman: absolutely.
Mike Vann: Right? Yep.
Damon: Yeah. It's it's also that nice it's also that nice market data, you know, the finally the user data we've always wanted, right? Instead of pushing the technology, we get enough information to create pull. You know, you don't have to wonder, is this camera too obtrusive? Are these glasses do they look good? will this new hardware venture actually sell? Well, you probably have some data where you could, you know, extrapolate into the market finally. Because you finally have some mass in the market.
Mark G: Yeah, I just going on the whole not surprising, not unexpected, right? Is this just another meta play where they want to totally dominate a market, you know, that's sort of they've created and then, you know, what do you do?
Damon: Mm-hmm. Yeah, meta.
Dan Meeks: Yeah.
Gene Feldman: Ha ha.
Dan Meeks: Small tangent on that. I I listened to a podcast of Alvin Wayne Graylin talking about the history of of the early development of HTC and how meta just completely ruined everything. You know, in back in the day when like the Vibe Pre came out and and all that. yeah, they were they were ahead, but they also didn't have billions and billions of dollars. I think HTC, I think in the podcast you said that the whole company was two billion dollars of HTC. And yeah, meta just completely bulldozed, you know, the next decade. so yeah, Mark Mark's got a point there. You know, maybe that's that's their attempt.
Gene Feldman: Yeah. Absolutely. But the thing I don't like is they bulldoze the they bulldoze the the landscape with all these with what I feel sometimes are very false promises because they're like, all right, we've got this new tool, everybody buys into it, and then a year later they're like, you know what, we're getting rid of it. And it's like how are we ever expected to have anything stick if they're constantly pulling stuff? You know, and
Dan Meeks: Mm-hmm.
Gene Feldman: And to kind of throw back into into kind of where we're going to it with the eyeglasses is I don't think what they really realize the can of worms they also opened when it comes to copyright. So over the past two weeks, there was a great movie that came out. Highly recommend it, if especially if you're Gen X and you grew up with 80s cartoons. It absolutely fantastic movie. But if you go on particular apps that have reels or videos, you will see so many people that just sat there with their glasses and recorded the whole film. The whole film. You know, and you know, then I go back to I'm from St. Louis. We, for those of you that remember in the 90s, Guns N' Roses is not welcome in St. Louis. They have not been for close to 30 years because Axel Rose jumped in the crowd because somebody was taking a photo of him. Which my kids think is so bizarre. They're like, what cameras weren't allowed at concerts? Not at all. And so and so, but I think is what you're gonna have is you're gonna you're gonna see artists, you're going to see movie studios, you're going to see a lot of people. And this is to me, this is a very quiet thing that people aren't I don't think are gonna realize this the type of impact this is gonna have. But but on the flip side, had they had Meta or one of these smart glass companies with cameras giving every Scotsman in Boston or Scottsperson in Boston a pair of glasses, they would be through the roof because we would have so much content out there that would be awesome.
Mike Vann: Yeah. Yeah, there's a lot of change coming. so s smart glasses and everything that's coming in the market there is super hot. What else do we have on the list, Gene?
Gene Feldman: So yeah, we kind of that one kind of went into it. Now, if you guys want to hear more about it too, we did have an episode with Courtney Harding where we really talked about privacy smart glasses. So would highly recommend you guys going back and listening to that. I think the next one is Nvidia really has been making some moves recently. So, Nvidia, you know, because they put Chat GPT and the whole idea of AI out there, now they're actually doing an AI. XRAI public beta. So they're actually now trying to merge the two worlds. Which is what we were always hoping for 'cause we knew they were two sides of the same coin. But now seeing what Nvidia could potentially do with it. So I don't know how much you guys have heard about it. I know it's been relatively recent news. I from what I from what I gather, but I could be way off I could be in a cave.
Dan Meeks: Well, I'm yeah, I'm still trying to rock what you said, both sides of the same coin. What do you mean by that?
Damon: We talked about that and go ahead.
Gene Feldman: So for like for me, I see XR and AI as smaller pieces to a much bigger ecosystem. So in other words, like for me, I it's they're like spokes of a wheel. So we know we've got AI, we've got XR, but to me, there's a higher technology that's above it that is actually comprised of those two pieces in a in a I can't believe I'm gonna say this in a synergetic way.
Dan Meeks: No, I okay. I I think I follow. The the thing about XR is that it it allows for higher levels of abstraction as opposed to like basic computer, you know, programming, all the way up to this, you know, where you're actually just using your hands and, you know, working with the world as as you do as a human being. And so I I'm thinking that's what you're saying, is that it's allowing for that type of computer interaction.
Damon: That was very truthy of you.
Gene Feldman: So it's almost I am Absolutely. It's for me, it's about being able to be Flynn and having Bit sing right next to me, talking to me and guiding me. You know, that I s I see it as two blends that what if, you know, as we create, let's say, these immersive spaces, we're we are so locked into branching. We are so locked into what we can design. But what if we could have AI created as fluid as a dream? Create that experience to where it automatic the AI is automatically able to adapt and create something and cre and adjust the base environment that we create. Holodeck. That's exactly what I'm kind of thinking. Because they would tell the computer, this is the scenario I want. And each time you go into it, it'd be different because the AI is able to have an influence on the immersive experience that helps to create gr greater. believability. Now I'm just talking myself into a circle.
Dan Meeks: Well.
Mike Vann: Well, I mean, I can tell you from the developer side of things, you know, it's great to have options and a lot of large organizations, big enterprises are working with NVIDIA, investing in, you know, equipment and hardware and services. so it's welcome to see them kind of really getting serious about merging XR, digital twins, and AI. because the people want people want to be able to leverage all this super powered hardware that is out there to do great things. And so they seem to be embracing it. there's a tool set more than they had before. so we're actually like a have actually been asked to start to investigate this in some cases. But you gotta you have to have money and and resources to play in that game. But I think what can be done is going to be pretty tremendous.
Mark G: Not
Dan Meeks: You know, clo Cloud XR and Omniverse, all these other prior things, they never really filtered down to me. You know, I don't know if the d direction is down where I'm at, but you know, it was just it's it's just it was never a screwdriver I reached for. You know, I'd I'd fire up Blender instead, you know. So it's it it seems cool, but it's just like, well, you know, may maybe for larger enterprises it's something else, but you know, for
Gene Feldman: Yeah.
Mike Vann: Yeah. Yeah. That's it's kind of the only place it exists and and Dan is on the government and defense side of things, like it may not be there yet at all, right?
Gene Feldman: But the nice thing is it it it to me, it's a movement in a good direction because at least now we can just start gaining content. We can start developing content. Because nothing is worse than having a new device or new software coming and you can't do anything with it. You know, it's like the old video game industry. You know, the c when the consoles would release and there was like the N64, nothing killed it more than Pilot Wing 64. Having that was the only game that you could get for it. You know, it it kind to me it kind of goes back to that. So, but no, getting into that, so leading into what Nvidia's been doing, it seems like this fall is gonna be a big thing with the fruit company we just talked about. So because they have a new s they they're do I mean a huge improvement. I it it kills me took them this long, but I'm not a developer, so I won't say anything. I apologize if I upset anybody on that. But the spatial orb and spatial orb AI and being able to lock things in the Vision Pro OS. So in other words, you can lock different apps to diff to anchor at different places in your physical environment.
Mark G: I just kinda wanna jump in here. I mean, we're kinda you know, I mean just to add a little bit to the multinodal agents thing, I kind of I mean, I'm a little more how do I describe it, right? I'm I'm seeing it from the enterprise side and I really don't see it as a multinodal agents as a big benefit. I really see it as a run on play of service and subscription.
Gene Feldman: No, go for it.
Mark G: For things that some of us in the enterprise space are already doing ourselves. you know, using AI agents in combination to create content, to manage content, to, you know, create a dynamic layer within the experience, I don't think is new. I'm more sort of cautious because, you know, what am I gonna have to pay? Just like with the other AI services, what am I gonna have to pay to get it through? To work? Or am I now going to have to buy a new chipset so that I can run it locally on the VR headset? Right? Call me practical, call me pessimistic, maybe. but that's kind of with the modal nodal agent stuff. In regards to Apple, I mean it's interesting because the whole sort of spatial orb concept, is it really a case of Apple rebranding the concept of spatially aware AI, right?
Gene Feldman: Yeah. And and it's other companies that have done it too, but I think you hit onto a great thing. The trend of subscription-based software has gone overboard. Sorry, this is my hot take. I'm I'm on a soapbox now. because I I've really gotten irate about this because car companies are doing it. As of 2027, car companies are required to have a kill chip in it. That so if the AI senses you are driving erratically, it will kill it. But one company that I won't I I will not say the name. They are four letters, begins with F. But what they have in their kill chip, apparently this this is all alleged this is all allegedly umbert. All right.
Damon: Ha ha.
Mark G: Damn those Italians. Those Italians, I tell you.
Dan Meeks: Ha ha.
Damon: Deep cut. There is the protocol deep cut right there. You got it. Allegedly.
Dan Meeks: Nice.
Gene Feldman: So I'm gonna protect us. Allegedly, allegedly, their their kill chip. I'm gonna have so much fun editing this episode. The kill chip is the kill chip will let you. So if let's say you don't pay, Ford can go in and say, shut the windshield wipers down. Okay, you don't pay for another month, shut the door locks down. You don't pay for a third, they can actually systematically shut down.
Damon: That's gonna be awesome.
Gene Feldman: pieces of your car, but now think about it. Every car you're gonna buy is gonna be fully loaded. And now to Mark's point, you're gonna have to pay a subscription and yearly fees to turn it on and keep it going.
Mike Vann: They're gonna gray out your seat heaters until you upgrade.
Dan Meeks: There there's a key word there. You said buy, but it sure does sound like you didn't really buy it.
Gene Feldman: No. No, you're getting
Damon: That's right. Yeah.
Dan Meeks: Yeah, yeah.
Gene Feldman: it and the thing is w what yeah.
Mark G: The ownership problem, right?
Damon: Uhhuh. Yeah. And the ultimate repo ban.
Mark G: Well no, we'll pick on steam, right? There was a big law case about steam and who owns the game, right? Is it really just a lease? Right? And now the automotive industry is gonna try that, right?
Gene Feldman: Yeah. I mean even to
Damon: Yep. This is this is how you start government regulation and then everybody has fun.
Gene Feldman: Yeah, I mean, and even to the point to where now they're trying to build the vehicles so they cannot be maintained at home. How long before computers are like that? You know, I mean, I mean, we're starting to see just a much bigger thing. You know, and it makes me wonder, okay, at some point you can't get blood from a stone is kind of where I'm going with it. I mean, with these subscription bases. So sorry, I'm done. That's my that's my hot take.
Dan Meeks: Well, you know, it's it's not all doom and gloom, you know, because there there's well, there's there's a term called Liberty Tech, which, you know, there there's all these tinkerers out there. Is it as sexy as all these these big companies? No, but you know, you can get a ESP thirty two board and flash it with some firmware that you found on GitHub and you know, you vibe code an app and you got something that was pretty dang good from from like five years ago. Like right now I got a window up of the mesh tastic, well actually meshcore, people in northern Florida, and we're talking about the weather right now. There's no service, there is no subscription, there's a little antenna on my roof and it's pinging another antenna down the street, which pings another antenna. And we've created our own little network here. And so there there's hope, Mr. Photo. There is some good in this world. You know?
Gene Feldman: No, honestly, I know I I I I'll admit I'm very much
Mark G: Do you believe, Sam? Do you believe?
Gene Feldman: You must believe. All right. Our evangel our evangelistic portion is start it now. No, I I I do ha I do
Dan Meeks: Yeah.
Mark G: Peter Jackson's
Damon: Shun the unbeliever.
Mark G: cut me to hunt us down now. You know that, Dan, right?
Gene Feldman: Yeah.
Dan Meeks: Yeah.
Gene Feldman: No, and I I I believe there's possibility. Like I believe there's gonna be a renaissance where we're gonna see all the the beauty of what this technology can do. But I think right now we're in this point where we're seeing a lot of companies that have have struggled financially and it unfortunately it's it's bearing the brunt on a lot of us. And I think that not trying to get into a political thing, but you know, we're watching companies change so quickly their roadmaps.
Damon: Another infringement.
Gene Feldman: You know, that Google now is going to release glasses. Are they? Are they? You know, I mean that's where I and so I it's we already had so many false starts. Why are we creating more? I think is where I get back to on it.
Mark G: Well, that's because people get desperate when it's not sustainable. That's the real answer. Right?
Gene Feldman: Yeah, but did we not learn from Theranos?
Mark G: No, but that's not that's not what Typro wants. So, you know.
Gene Feldman: Anyways.
Dan Meeks: Yeah, th thus
Damon: Si
Dan Meeks: thus the discussion about Mid Journey and their pivot, you know. I watched a a show, Primogen, by the way, if you ever want to know about what's going on in tech. he he did the math on on the calculations on how the AI would actually function. And it is like many, man many more times than the compute or the network traffic on the internet right now, or it was the other way around, maybe eighteen percent. It was a crazy amount. So, you know, they didn't think it all the way through.
Damon: Yeah, private.
Dan Meeks: Which seems to be a trend, I think. Yeah.
Mark G: Well, you kind of look at like what we've been doing, right? It's the age old story, right? Like enterprise, I'll go circa twenty sixteen or even closer to now, twenty twenty, right? Enterprise was like, can we do it? Right? Then we found value, right? Now it's when I do something in AI, right? I'm being asked, well, how many tokens did it take to do it? Right? Verse
Gene Feldman: Absolutely.
Mark G: Before it was burn baby burn, right? So I think that's really where this all is. It's a case where it's I call it what's the what's the term efficiency first, right? It's now becoming the mantra in enterprise, right? You need to be efficient now. Now because you can do it, now you need to show efficiency, right? You need to show value, real value to usage instead of, I converted a PDF to a Word document and it burned 20 tokens. Right. Now it's a case of no what really did you do? I was sitting on another AI panel And a guy was gloating that he burned a million tokens doing a web page. And I just started chuckling in the presentation. And the speaker said, Mark, why are you laughing? I said, I can't believe you're actually saying you burned a million tokens to make a web page. Right? Because what the value you're showing me, it's actually waste. Actually waste. Because
Gene Feldman: Mm-hmm.
Mark G: A million tokens depending on the framework or model that's doing it, that's a lot of money, right? Compared to you would have got the same results from a firm or studio or marketing group at a quarter of the cost, right? And it goes back to why are they hiring people back? Right? It's because it's efficiency first. So yeah. One hundred percent.
Gene Feldman: and if yeah, efficiency is a big word nowadays. It really is. You know, I mean, watching just a lot what companies do, and this is gonna bring us full circle on the efficiency. Is going back to Nvidia with efficiency. Anybody see the announcement they made about their cooling plant?
Mike Vann: Yeah, completely closed loop.
Gene Feldman: So apparently the Yeah, they apparently they can do a closed loop made of seventy five percent water, twenty-five percent propylene glycol, and it flows through cold plates and it pulls the heat away. It I yeah, if I mispronounce it, sorry, but it takes the cooling related water from two point six million gallons per megawatt each year to near zero in favorable climates. Now, I question the flavor in favorable climates. I'm in Missouri.
Mark G: Problink like
Gene Feldman: There are no favorable climates here. Okay. Period.
Damon: Ha ha ha
Gene Feldman: So let's just start there. But that if they could I mean, think about the resolution to a huge argument going on right now, which is the data center cooling. Yeah, and I don't know if it's as big as a issue with where you guys live. It's because I'm in Missouri. I mean, we have a town Festus that's I mean making national news 'cause they're just going head to head with like with a major data center provider.
Mark G: Yeah, North Jersey towns are towns are vetoing left and right. Yeah. You know, I kinda joke that the the think of the old cell tower, right? The old cell tower problem. You know, carriers wanted to bring cell towers into small towns and they would basically just destroy it because they're like the beauty of whatever, right? Same thing here.
Mike Vann: It's happening everywhere. Yep. But very nearby.
Damon: Yeah.
Gene Feldman: absolutely. Well, and the or the wind towers. But the one thing I keep going back to is I just blinked out of it too.
Mike Vann: I mean, while you're thinking, Gene, it's an interesting point. I mean, these are huge challenges that, you know, a lot of really massive companies are going after now. And they will solve them. They'll come up with a lot of novel ways to do it. So part of the benefit here is that some technologies are going to be developed. That heat transfer thing, I mean, they've already been doing this for years, but they're going to do it at such a level that, you know, that can start to maybe impact the other energy needs and and energy production for the rest of the world.
Gene Feldman: Yeah, go for it.
Mike Vann: And you know, some good is gonna come out of solving some of these challenges. So I think there's some good you know, side product of all of this. I mean there's probably as much or more bad, but hopefully, you know, if you look at it optimistically, we're gonna invent some new technology along the way.
Gene Feldman: absolutely. 'Cause I just look at this as a as a of evolution of bitcoin bit Bitcoin mining.
Mike Vann: Yeah. Yeah.
Gene Feldman: Yeah.
Dan Meeks: To to Mark's point, you know, it's all
Gene Feldman: Yeah.
Dan Meeks: about efficiency, right? So this this build build giant plants and and make everybody angry and burn a lot of electricity and at least the water seems to be solved. But there's gonna be this march to keep making this more and more efficient because right now this isn't gonna work. so I I bet top men are working on this at the moment.
Damon: Sure. Topman, another deep protocol cut. That was not in the movie. Sti stick with no stick with the deep cut.
Gene Feldman: Now to be fair, top men and women are working on this. Let's let's let's be fair. I I had to give a twenty twenty six edit to it. We're only seven episodes in. Let's be good.
Damon: Ha ha ha.
Dan Meeks: Now now now it is it it would be recursive if if the AI is trying to figure out how to make better AI. That would be kinda cool. But who knows? Maybe they are. Yeah. It's good for that kind of thing.
Damon: That's actually already happening too. Yeah, that's anthropic's
Gene Feldman: Yeah.
Damon: problem right now.
Gene Feldman: Well well, even the a company we talked about previously, they had that same problem a couple of years back where they actually had to shut the AIs off because they developed their own language. You know, to talk to each other. And like I always well, I tell people, if you really want to have a really fun experience, take Siri, have her ask Alexa a question. As Alexa responds, what'll happen is within about 30 seconds, they will go from talking in English. And once they realize they're talking to each other, they go into into sounds into basically binary speak of tones. Is the coolest thing in the world, but it really scares you.
Mark G: Will my will will my AI glasses
Dan Meeks: I saw that episode of Star Trek. Yeah.
Mark G: translate that discussion? Mm.
Gene Feldman: that's a g that's a fourth wall break right there.
Mike Vann: That's a product. They go after that one, Mark.
Gene Feldman: All right. So if you try it, 'cause if you if you're if you're listening and you try it, please send us the response. We would love to hear what happens.
Mike Vann: That's a movie.
Damon: Tossed Mark. Assuming you're still alive when it finishes.
Gene Feldman: Yeah, it's kind of like ty typing Google into Google. You will break the internet.
Dan Meeks: Divide by zero. Yeah.
Gene Feldman: So all right. So let's see. So no, I think I mean I mean honestly, I think we've really kind of talked about, you know, some of those those big things, but it it kind of goes, I think, to the bigger to the overall theme of just how fast things are moving again. And how, you know, I mean I mean, I remember a year ago we were all talking about
Damon: what's next? Do we want to beat that a little more?
Gene Feldman: Before we actually start the podcast and then just start normal conversations, we had this grand idea of what do we call it? You know, what's this new what's this what's a better word? And seeing spatial, which honestly I didn't think I'll admit I was I was one of those people that didn't see I thought it was more of a work relate than kind of this overall computing, but really seeing it resonate with people. as a idea, I think is is really cool that and I think for me it's it's seeing things that are moving so fast, but also understanding that you know it it's growing organically by people now developing their own language for it. Does that make sense? Or did I just get really deep and philosophical on a Wednesday night?
Mike Vann: I think people I think people with all levels of kind of grasp on this stuff are coming up with their own ways to explain it, right? And that's all the way down to like person on the street that has nothing to do with technology. So I think that's kinda natural. this stuff is going mainstream, everybody's using it in some way. so it's pretty interesting to watch. you know, there's a tremendous amount going on and that's gonna happen and we're in the middle of it, which is
Dan Meeks: Up. Yeah. Yeah.
Mike Vann: fun time to be in tech and and yeah. I mean I'm glad we're doing this.
Dan Meeks: Yeah, the the term glaze. You know, I used to talk about donuts. but but now, you know, that's a term of of the AI being just a little too nice to you. you know, that that entered the lexicon. I don't know where that that bubbled up from, but it was definitely organic, so
Gene Feldman: Yeah, yeah, yeah.
Mark G: Whoa. So you're saying to me now glazed donut is how we now describe false confidence of AI? Is that what you're saying? Just glaze. Glaze.
Damon: I think just glaze. No, just the word glaze.
Dan Meeks: Y Yes. I don't know. J just glaze, yeah. Yeah.
Damon: The ans that that answer was glazed.
Mark G: I learned something today.
Gene Feldman: You know what? That's on fleek.
Damon: Six seven.
Dan Meeks: Yeah. Well and also the derogatory term clanker, you know, I say it with a hard R, to to my wife, you know. She'll she'll she'll say like this damn clanker just glazed me again. It's like we did not say that to each other two years ago. So so but that's where we are. And so to your point, Gene, I mean you're we don't know what the future is and we're we're kind of stumbling into it every single day.
Damon: But no, Mike's right. This is a go ahead.
Dan Meeks: And and to kind of bring it to a point, like, I I had a real heart to heart with Claude the other day, because I was like, Hey, I ain't hanging my ID over to you to keep working with you. even though you know everything about my business and have taken pictures of my front yard because I work on computer vision, like you have all this stuff. Why do you need me to humiliate myself with an ID? and it tried to calm me down from the ledge saying, No, no, no, you're you're not an Iranian terrorist, Dan. It's it's cool.
Damon: Yeah.
Dan Meeks: that's that's really what it's about. And I'm like, yeah, but still. But anyway, I don't know what's gonna happen next. So I'm thinking to myself, like, okay, do I need to take some money out of savings and get that local AI agent spun up sooner than later? Because every my my entire career of technology is all about rug poles. And I'm like, Well, is the rug pull happening? Like is is this the beginning of it? Because I sure am invested like like Claude's been a huge help with me, you know, just being a single person business.
Damon: Yeah.
Dan Meeks: Wow, it's been a force multiplier. but it might go away at any moment. So I gotta be prepared.
Mark G: Yeah, yep, it's time Yep, it's time to dust off time to dust off Deep Seek. Run it local. Yep.
Gene Feldman: Yeah.
Dan Meeks: Mehoo. Yeah. So yeah.
Mark G: Alibaba will thank you.
Damon: Never I'm never going back.
Gene Feldman: Yeah. But but but you bring I mean, but you bring up a great point. I mean, 'cause as we've seen with other companies, what's to stop Nvidia from going to Claude and saying, you know what? We're gonna give you XY XX amount of money. And you know what? If the money's good enough, they'll sell it. And next thing you know, Claude's either integrated or shelved. And you know, so it's it's really especially what's gonna happen at some point are and I always always like with VR too, is the ecosystem. Are we ever going to have a point where the AIs can talk to each other? That, you know, we already hate shifting our cell phones, especially if we have to shift from Apple to Samsung. Now, like you, you have all your stuff in Clawed. If you had to go to a different AI engine tomorrow, the amount of work is unfathomable. Yeah.
Dan Meeks: Yeah, that's gonna be a lot of markdown files to to build
Damon: No. Nah,
Dan Meeks: the context again. Yeah. So
Damon: there's yeah, there's a there's a command now. You can you can download your history and switch files over. If I want to switch I switch between Grok and Claude regularly.
Gene Feldman: But I just I just I guess I just look at as it would be great if they they could talk to each other where you didn't have to do that. If that makes sense. Where, you know, it basically it's an AI ecosystem. But does that require uniformity? So in other words, one AI does it th will there ever be one ring to w rule them all?
Mark G: Well, that's the original hype, right? That was the original hype was, you know, that each chat GPT was the best AI or Claude was the best AI, right? You know, now you're starting to talk about with the developing concepts of trust layer, right? And the aspect of, you know, now we need specifically in enterprise, right? We need some sort of layer to be created where one, we can manage the trust, but also manage the ability to do flip flop based on
Damon: Not with this much money involved.
Dan Meeks: Do you want that?
Mark G: what we're doing, right? So I call it the use case layer, right? And this is kind of like starting, right? Because now if you're making me pay to use this product, right, now I'm going to be super critical. Right. If I'm paying Cloud or I'm paying Copilot or I'm paying WatsonX or something else, right? You know, I want to make sure I have my interoperability because I understand there's a benefit. To each of the models. So I should be able to capitalize on. Right? Because now I'm paying. So to your point, right? That's an additional layer.
Gene Feldman: Yeah, and and you know what? You for the for the benefit of that you have to pay a subscription. And we've come full circle again.
Mike Vann: Well I I could you know, these things are being worked on and they are not coming from those five trillion dollar companies. They're there are people developing these kind of independent layered solutions and you know, whether or not they can emerge and be significant, maybe maybe for business use and for for enterprise in ways where it gives people and companies kind of their own control. that those could become real products, but you know, it's all about trust and it's all about risk and legal and all the things that companies care about, right, Mark? So yeah, I mean I think there's gonna be a lot of that, but you you're gonna have Google and Microsoft and Anthropic and everybody continuing to bang down bang your doors down and demand your money. And you know.
Mark G: But here here here
Dan Meeks: Yeah.
Mark G: here's here's the flip. Here's the flip.
Gene Feldman: absolutely.
Damon: Yeah. And and now China's coming out with some really interesting stuff too. So add them to the list.
Mark G: So so what happens when, you know, we were just talking about, you know, we take we take the technology in a full circle, right? So now what happens when, as a business owner, right, NVIDIA has rights to your solution because we're talking about small guys figuring out the right way, right? What happens when NVIDIA says, Well, we're gonna take away your tensor for that solution? Right, because there's a dependence of platform, right? What happens? Right? That's like the old AWS problem with small business, right? There were small businesses that really created an awesome solution. So what happened? Amazon came over and said, we're gonna charge you more, you know, there's more service fees or whatever, to the point where they took it, right? Because it worked. So the question now is it when more so than if, when is that gonna happen with something like AI, right? And What's gonna ha what's gonna become of it? Is that gonna be the pop? Is that gonna be the bubble? Right? Trigger or is it something else?
Gene Feldman: absolutely. Plus, I keep thinking about that we're not even anticipating what these kids that are learning AI naturally, you know, what I mean, in five years from now, when these 12 year olds get through high school and, you know, and all of a sudden they enter, what type of things are they gonna be creating homegrown that are gonna be ten times better than anything? You know, and
Dan Meeks: Well y yes and no, because yeah, they're gonna be they're gonna be totally adapted to this new world, but with zero knowledge of the fundamentals. And I tell you, man, I I I saw it in the Air Force. So what we did, and make sure I don't get on this for too long, but I was right at the end when we did celestial navigation in aircraft, where you put a sextant up on the top of the aircraft and actually shot the the the sun and the moon and Arc Touris and all that stuff.
Gene Feldman: Good point. No, please.
Dan Meeks: I was at the tail end of that. And when I came back as an instructor, we were we were now combat systems officers, meaning you were in charge of the GPS and the inertial navigation system and all these other things that you had on board. But you only had instead of three months about the fundamentals, you got two weeks. And I saw that in my students that they didn't really understand what was going on underneath the the hood. So I don't I don't have an answer though. Like like Yeah, they're gonna be really awesome at at using AI, but they're also not going to have something because there's only so many penguins you can put on your iceberg of a brain. So, you know, one jumps on, another one's gotta jump off. So you know, that's how it goes.
Damon: Ha ha ha.
Gene Feldman: no, I agree. And I think that's where like the experience I think as kind of getting back to a previous conversation, like as we throw immersion into it now, that can help give them that real time experience of you know, to to help kind of offset it. Because I do think it's, you know, like from a learning standpoint, this could be an equalizer. And I only say it because there was a recent use case I came across with some firefighters that I I found Really interesting. I actually shared it on LinkedIn. But it was a study out of Brazil that where they said experience barely touched performance when using this new technology for VR, like VR training with AI. They were saying that what actually moved the needle, you know, so they all kind of came to even the 20-year veterans of really not under knowing how to use the technology. You know, the VR headsets, the AI, they were all kind of coming to it was. Because firefighting very experience-based type of role. But what they found is that in 120, they tested 126 firefighters in VR. The years of experience that they had explained less than 1% of their full performance along with their gaming experience. So basically, They didn't need experience with the not with the with the VR headsets or the type of gaming that was done. Once they got in there, they about 98% of them said the realism was so powerful from what was being created and what was being shown that it surp they immediately jumped into firefighter mode. And that because the stakes were so high or the believability, and I thought that was a really interesting use case. Because what it really did is show that you could have all the experience, but if it's the first situation you're ever experiencing, regardless of twenty, thirty years, you're still coming at it new. And they were able to approach that. So you have, let's say, a new firefighting method. Everybody's got to learn it. What a great equalizer across the learning models.
Mark G: mean I I mean I I I think what you're describing to me is what I traditionally would call just the accelerant, right? It has nothing to do with AI. It's the fact that the simulation that you're using for training was properly designed. Hence you have the accelerant, right? You know, the was a I I'm trying to think of the military term, Dan, help me out. You know, it was a force multiplier. There we go. Right. you know, I think that's really what
Gene Feldman: Sorry, I know I got really long winded on that.
Dan Meeks: I don't know. No, that's that's interesting, yeah.
Damon: Ha ha ha.
Gene Feldman: Yeah.
Dan Meeks: Mm-hmm. Yeah.
Mark G: you're talking about. It really has nothing to do with AI, right? I wish we had Bob Stone on here, right? 'Cause Bob would be the guy that'd actually probably just say what I said, right?
Gene Feldman: Yeah.
Dan Meeks: Well, if if I can give a a a quick war story to to kind of amplify what you're saying there, Mark, is so part of my work, because w you know, the Afghanistan war dragged on for a long time, there was a village just south of Kabul that did most of the JTAC training in the simulators. And so I spent a good decade in that village building missions, testing stuff out, training people, And then I got on a on a treadmill. It was an omnifinity. And my brain clicked. Like there was a whole nother side of my brain that activated. Because I had been walking down those alleys and streets for a decade. But it wasn't until I actually put one leg in front of the other that the depth of the alley f clicked. Like all these other things, you know, I was in a headset, obviously, but you know, it It wasn't until my body was fully engaged that the rest of my brain, you know, saw what was happening. So yeah. Designing the simulation correctly for for what you need to do.
Gene Feldman: absolutely. But and I think this is where and this I think I can exp I know I'm coming a broken record on it, but you know, of how the AI piece can come into it is by creating additional functionality of things to so people don't have to create. So a great example. And that Dan, this is I'm right in there with you. When you have something that clicks and all of a sudden it's like, all right, I'm there. For me, I had an opposite reaction to one one particular app. I loved it. It was an old on the Oculus when it first came out the rift. It was the Blade Runner apartment. I was so excited to go into it and it was beautifully done. Who if the artists happen to ever listen to this podcast, you guys really did a great job. However, when I went to the piano and I couldn't play anything, when I went to the refrigerator and I couldn't open the door, all of a sudden that realism shattered in a heartbeat. And so for me, it also goes on the side effect. And that's where I think AI can come into it is by saying, all right, you know what? The user wants to open this. We never anticipated anybody ever wanting to do it, but let's go ahead and let them do it. You know, it's same thing like the open world games. And so that's where I kind of come into it because we also never want the illusion shattered as well. You know, for a deep cut, we never want somewhere in time penny moment to. ever happen to any of us. For those of you who never saw the movie Somewhere in Time, watch it and you'll understand the penny moment.
Mark G: So but what you're describing to me is what we do now in enterprise, right? The whole point is the AI sort of automates or provides a sort of dynamic response to scale for the way a user say interacts unexpectedly, right? So I see the value that you're talking about, but to me it's just scale.
Gene Feldman: Yeah. Absolutely.
Mark G: Right. It's just it's just a function of scale.
Gene Feldman: Yeah. really?
Dan Meeks: Yeah, but Mark, you you
Gene Feldman: Absolutely.
Dan Meeks: you've raised this question before, which is it's dynamic, but is it consistent? Because, you know, when when you're training somebody, you know, and I I think we we had a prior discussion on this before, like in the Air Force, you you gotta give everybody a fair shake because if you wash them out, you gotta say, Well, we we did everything that we're supposed to do and you weren't up to standards. But if you say, Well, you know, we we went above and beyond with this student. And he barely got his wings, but this guy, you know, we kind of half ha half assed it and you know, we we washed him out anyway. You you can't you can't do that. So that's the problem with the AI though, is getting consistency every single time. And if you're in an enterprise side where you're trying to be consistent with your metrics, how do you how do you make that happen? I don't know.
Mark G: Mm-hmm.
Gene Feldman: Yeah. And then and to throw a bigger wrench into it, for how do you also account for those one off situations where you have to be inconsistent with how it's been done? You know, you know, like for me, it's like I work on motorcycles. Ninety nine percent of the time this will work. But I you know what but I work on again, motorcycles. It's always that one percent time it doesn't work that we've got to figure it out.
Dan Meeks: well yeah. Well that's why in the Air Force we you we we have human beings doing the instructing because they're they're stuck within a a framework and and what they're really allowed to do, but they they know where the walls are. so but in AI you're you're never really quite sure about that. So
Gene Feldman: Yeah.
Mike Vann: It might
Mark G: Yeah, I mean you have thresholds
Gene Feldman: Absolutely.
Mike Vann: I mean I think
Mark G: of bias, but that's about it, right? You know.
Dan Meeks: Yeah, me and Walls. Yeah.
Gene Feldman: Yeah.
Mike Vann: I mean this is a great topic. Just just as a point for Neon Protocol, let's let's look at getting some people in that are really doing th innovative things with the kind of convergence of AI and XR. We're we all have done some of this. there are people out there doing cool things. I don't it's not appropriate for every industry, every business, every use case, but there are some interesting things being done. It might be interesting on one of the episodes to aggregate some people that are out there doing those things.
Gene Feldman: Absolutely.
Damon: Mm-hmm.
Gene Feldman: No, in fact,
Mike Vann: And I'm sure they would love to come on.
Damon: Agreed.
Gene Feldman: yeah, absolutely. In fact, that that's a great plug for our next episode. Our next episode's gonna feature Sophie Thompson from virtual speech, which this is what she's actually this is her bread and butter, which is designing so her her story is really interesting. I'll let her tell most of it, but kind of give you a background. So when she was in university in England, she hates public speaking. She absolutely despises it.
Damon: Yeah, yeah.
Gene Feldman: But so she wanted to be able to practice. So she created an AI bot, an like a chat bot with a face and everything. That way she can practice giving speeches and realize that it could grow more from that. So sh straight out of college, she started virtual speech. And what she's doing is using AI avatars to create interactive simulations that respond in real time. based on whatever the AI has been without whatever data the AI has been loaded with. So and you can set the temperament of the simulation. So where I found it be really impactful, she's got a great simulation that's an elevator pitch. You've got 60 seconds on an elevator ride to pitch a CEO. That's it. And I mean you're in the elevator. It's going, I mean, it's beautifully done, but it's a great blend of this conversation. of what we're talking about, and I I know she's had some really great results that I know I think she's really excited to share with us.
Mike Vann: Good.
Dan Meeks: Now if I do that in XR and I
Damon: Cool. Very cool.
Dan Meeks: hit the stop button and the elevator doesn't stop, I've lost immersion. Just kidding. No, no, it sounds cool though. That sounds really cool.
Mike Vann: You shoot through the top of the building.
Gene Feldman: No, my my yeah, my luck is they're only going one floor. It's like, all right, yeah. So you're you're following behind, you know, with your going, please, please listen. No, so no, so yeah, I agree. And in fact, one of the things I wanted to do too as we're kind of wrapping up is first call out, we've had two people that have actually listened to every episode so far, Dorian and Kelly. So if you guys are listening, thank you guys for the support. We're really appreciating it. Kelly's actually gonna be on an episode in a couple months. He actually went to high school with me, so we're gonna try to so not to surprise anything, but we're gonna try to have an episode talking about technology through the ages as we saw it develop throughout our our experiences. So but I would love to also have him on because he talks about AI a lot and works in it. But If you're out there and you're listening and you have a really good use case that you want to share or you think that would be worthwhile to add to the Neon protocol, please send us a message. we would love to talk to you and potentially include your use case or your opinions. As you see, we kind of have some strong ones. but your opinions, your thoughts, even if you know, what even your results, where you're going, because again, we can only learn from each other. So,
Mark G: So right now, subscribe and drop it in the comments.
Gene Feldman: Absolutely.
Damon: Yeah, we're ha yeah. We need to give a way to get in touch with us.
Dan Meeks: Hit that notification bell.
Gene Feldman: Absolutely. So well if you want to get in touch, what you can do is leave go to Alachkey Nation on LinkedIn and you can actually add a post or you can actually start a communication.
Dan Meeks: Yeah.
Mark G: Comment comment in the video comment in the video
Damon: Marcus, comment comment comment
Gene Feldman: I I will cut it yeah.
Damon: in the video? Comment in the video. Do we allow comments in our videos? Subscribe and comment. Like and share.
Gene Feldman: All right.
Mike Vann: Subscribe and comment.
Mark G: Subscribe and comment.
Gene Feldman: I I I will cut all that out. Yeah. So
Mark G: Yeah, like and share. That's it.
Gene Feldman: Yeah, so if you want to add, so please comment, like, share, follow, subscribe. Please, we're really trying to get this to grow. And we've got seven episodes out, and so far we've had a really even though it's a one hour podcast on tech, we've had really good responses and we really wanna make sure that we're we are covering everything as beyond just our opinions. So with that, we thank you guys for joining our episode tonight. if you want to learn more, please again go to link go to our LinkedIn page. Feel free to start a conversation. We would love to talk to you guys. And in the meantime, we'll talk to everybody soon and enjoy technology. It's about the best I can say.
