What happens when it’s faster to build a tool than to search for one?
In this conversation, we unpack how dramatic cost reductions and ease of creation are changing software, skills, and entire markets. Tool marketplaces are overflowing, discovery is breaking down, and the real bottleneck is no longer technology. It’s clarity, imagination, and timing.
We also explore why this creates a dangerous landscape for new businesses, how large players gain advantage through scale and compute, and why the next major shift is AI working side-by-side with humans as real computer operators, seeing screens, guiding actions, and transforming how learning and work actually happen.
This isn’t just about tools. It’s about how creation, competition, and opportunity are being redefined.
𝐄𝐧𝐭𝐞𝐫𝐩𝐫𝐢𝐬𝐞 𝐃𝐍𝐀 𝐄𝐜𝐨𝐬𝐲𝐬𝐭𝐞𝐦
𝐋𝐞𝐚𝐫𝐧. 𝐁𝐮𝐢𝐥𝐝. 𝐂𝐨𝐧𝐧𝐞𝐜𝐭. 𝐋𝐚𝐮𝐧𝐜𝐡.→ https://enterprisedna.co
Everything you need to master analytics, apps, and automation, all in one ecosystem.
At Enterprise DNA, we help you grow from learner to builder.
Explore structured learning paths, AI-powered tools, vibrant communities, and expert guidance, all designed to help you learn smarter, build faster, and launch better.
💬 𝐒𝐡𝐚𝐫𝐞 𝐲𝐨𝐮𝐫 𝐭𝐡𝐨𝐮𝐠𝐡𝐭𝐬 in the comments. Your perspective helps others learn.
📢 𝐒𝐮𝐛𝐬𝐜𝐫𝐢𝐛𝐞 for weekly insights on data, AI, and innovation.
Оглавление (2 сегментов)
Segment 1 (00:00 - 05:00)
It's a really non-obvious cost reduction, but it's sign it's so significant, isn't it? [clears throat] Just to actually do things that 3 years ago like we're talking 10 times less cost like very likely more like you know in terms of time and actual money to execute these sort of things like that democratization and reduction in cost. It's just spawning. It's just spawning just like so much new stuff that's hard to even fathom and imagine. But that that is what I'm I believe and what I what I think I'm seeing in a lot of what I'm personally doing. But also just extrapolating that out to basically every part of the economy. Um it's it's just like I'm just trying to think like a bit more high level like it's just absolutely mindblowing you know um what that is going to mean for humanity for us you know all the skills are like just this com like combination or culmination and all these many pieces of software that boom instantly insert into your workflow boom spawn something else. It's a evolutionary point of view. It's crazy, right? — So when the internet came on, the the big unlock was you could put all these things in one place. You know that all the shareware, you know, kind of march that came up where, you know, you could look for utility, it would be there. And then it got to be those got to be so overcrowded that um curation became really important. And then it was like you had to have trusted people to tell you which of these were good and which weren't. — And now it's gotten to the point where, you know, when Claude Skills came out, you know, I was like saying all we have to do now is we have to go um onto a skill mart, take the skill we need and pop it in. Well, the problem is they're so easy to build. In most cases, the skill marts are acrewing a thousand new skills a day. And so you're back to the point where — the mart is almost too big. And the interesting thing is it it's not curation that solves the problem. Now I just build it myself. It's faster for me in some cases to build a new skill than search through the mart and you know the six that are there and figure out which one is best. — I just say I build them for myself. — I know. Well then it just becomes your imagination which is holding you back. It's like knowing what skill you need basically. — Yeah. But the thing is even when you know exactly what you need, you're looking for it, — the search time is now in some cases exceeding the build time which is incredible. — Well, it's kind of like information and content, right? It's exactly the same thing. I mean, you're probably wasting your time looking through Google and searching through blogs when you can just like ask Chad GBT or ask Gemini or ask Ford now. Like, yeah, it's kind of the same thing, isn't it? Um I guess that's why Google have um put AI overviews over everything because it's it really you know the equation is so obvious now. So it's so obvious [snorts] um it's such a landmine or um [clears throat] sorry a minefield for starting new businesses isn't it because like one minute you think you've got this uh great AI business idea and the next minute skills shows up and you and you know like the amount of AI spreadsheet businesses that try to start up in the last like two to three years and skills and um chat GBT um agents are just like basically destroying those. I mean they they've got no chance now. Um it's there's that's just one small industry, — but um man, it's a it's an absolute minefield for the opportunities are great, but the um viciousness of capitalism is also greater. You know, it's also much greater. Well, and the you know the thing it takes to actually plow new ground um is the resources are enormous — that you know you think about the computer you know even something like nano banana you know which I'm saying something like you know which is huge — it's like who would have the resources — to do something like that fresh from the start other than like a Google Yeah, — you know, it's like that's not something a small business is going to, you know, even if they had the idea of exactly how to do it, they're never going to have the resources to match what it takes to put an idea like that in motion. And so, you know, you and I have talked I think the weird thing about this is the road for small business is just so difficult.
Segment 2 (05:00 - 09:00)
— Um, — in terms of coming up with something new. Um, and even if you find a really clever gap between what's offered, it doesn't mean that next week, you know, they're not going to just incorporate that into their portfolio. — I I'm just shaking my head these days because — one of the things it did in the midst of this is it told me when we went to the manual mode, it told me click on this, um, you know, click okay. And I'm like, that option isn't available. — And it said it should be there. And so I sent it a screenshot and it said, oh, okay, this is the problem. You're in the global um profile. The thing I'm talking about is only in the local profile. — And I'm like, okay, that turned out to be exactly right. But a, how do you know that? And b, um, the thing that I realized is at that point, and maybe I don't even know if Gemini has the ability to do this. If I can show my screen and we can work together in manual mode and it, you know, it can say, "Oh, no, you're in the wrong section. " Um, — that's going to be a whole another level of, you know, learning and education. — Yeah. Well, you just think about how like that they sort of turn them like computer operators, right? Like I don't even think that has even started yet. like um you know and just think about the computes that would be required like that's why the that's why you need all these data centers that are being built right because I don't even think I mean I don't even use the operator stuff myself and I'm a heavy user of AI so you think about anyone who's using a computer eventually they're going to want some sidebyside AI like looking at your screen just like working with you I I think that's going to be huge absolutely huge like I would love that I don't think it's just there quite yet like it's just not — probably because they can't facilitate it. The these companies can't facilitate the actual um the actual need. — Well, I think it was like three years ago. I was at a um Microsoft presentation and they showed they wheeled out like it was like a piece of machinery. I don't remember it was like engine or something like that and they had somebody wearing VR goggles and the person was working on the engine talking through the mic and then what they showed on screen was what the person would see through the VR goggles — and it could basically like overlay like a virtual lens kind of Google Lens — um — on top of the — the physical engine and show okay here's where part goes, here's what you take out. — And um you at the time, you know, it was clearly just kind of prototype stuff. Holo Lens, I think they called it. — Hol. Yeah, that seems to be that's one of the another ones in Microsoft graveyard at the moment. — Yeah. But the concept is phenomenal and I think it's going to be the way things work because as I say like we were so close in that discussion to basically that thing if I would have been able to show my screen in real time to uh to Gem and I think it could have just worked me through it in the exact same way. The unlock for virtual reality which has always um had a lot of potential but never reached its potential is the world models. That is the unlock isn't it? Is that it always was cost prohibitive to build these worlds and if you can just render them on the fly with AI um that that's what's going to unlock virtual reality. I I truly do. Um — that is a better form factor. — Yeah. Whether virtual reality is a good thing for humanity, I'm not sure. I'm really not sure. Uh that but um kind of like social media, but um but that is the that is the unlock. And just think of the compute required just for that. I mean that's just insane. There was this there's this really interesting um YouTube clip that I saw of um the top um guys at Microsoft were with a podcaster who's become quite famous um and they were walking around one of their new data centers. I mean, this [clears throat] is the most incredible state-of-the-art facility you've ever seen, you've ever seen. And um one of the things that to satella joked, I I suppose, aren't I meant to be running a software company here? And he's got this like beautiful facility of like, you know, these racks of Nvidia chips and stuff. So yeah, it's classic. Absolutely placer. I mean, they are going to be a physical asset company, man. As I've been saying for a long time