Ron Close: Marketing in the Age of AI

Research by Ron reveals how AI (ChatGPT, Gemini, etc.) ignores conventional sources such as business to business publications and industry analysts, instead citing vendor pages, Reddit, resellers, and even mystery sites. Guidelines proposed on how AEO (Answer Engine Optimization) might work in practice. 

• Comparison reviews may be the clearest path to being sourced by answer engines
• Ron’s AEO research goals
• A prompt-based testing method across ChatGPT and Gemini 
• Missing citations and why vendor or reseller sources get treated as authoritative 
• The CADSoft USA mystery and how high-volume comparison pages get rewarded 
• Why advertising incentives weaken trade media independence and credibility 
• Zero click answers from search engines and how AI previews reduce website traffic 
• Comparison reviews as AEO 101 and the business risk of publishing them 
• The adjacent CAD ecosystem and how LLMs pull in simulation, PDM, and visualization tools 
• The opportunity for a Consumer Reports-style authority for CAD and engineering software 

Transcript

Welcome And Quick Catch Up

Roopinder 0:00

Welcome to EngTechnica TV, where we bring technology into focus by talking to leaders with technology of interest to engineers. Hey Ron, how you doing? I'm doing well, man. How are you? Good. I was just finished my coffee, but you're probably having uh your evening cocktails.

Ron Close 0:18

I was thinking about cracking a beer while we spoke, but then I'm not sure people would know what time zone I'm in.

Roopinder 0:24

So well, you're welcome to one of one of our podcasts is actually called uh happy hour. And I encourage I encouraged drinking on that show.

Ron Close 0:33

Oh, that's funny. I haven't got the invite for that one. That one sounds like more fun, man.

Roopinder 0:40

Yeah.

Ron Close 0:41

Hey, I just returned from nTop conference. I saw your note there.

Roopinder 0:45

I knew them um when I worked at ShapR. Let's look at those two as the what should I say? Leaders in the next revolution of design.

Ron Close 0:53

Disruptors, yeah.

Roopinder 0:54

Yeah, disruptors, disruptors. That's right. That's right. So Ron is living in Budapest, Budapest, I should say. Again, I think you tried to get out and they wouldn't let you back in the country, right?

Ron Close 1:08

Well, that's a long story, maybe for for another discussion. Um, but my wife uh was not able to come to the US. We decided to settle here. Yeah.

Roopinder 1:22

I think your wife is off not of American birth, right?

Ron Close 1:25

She's uh Romanian passport. So um uh so I'm going to be an official European resident in the next week or two. Hey, congratulations on your election. We walked down to the rally uh for Peter Magyar, and um it was it was just the the buzz there was crazy. As we're walking across the bridge, cars were slowing down and giving high fives to people walking by, honking horns, hundreds of thousands of people out there. It was it was uh, and since that's happened, there's just been this incredibly positive vibe here in Hungary. It's uh it's amazing to see. Absolutely.

Roopinder 2:05

I'm great. Something similar to that comment.

Ron Close 2:08

Yeah, yeah, yeah. Yeah, I think that's kind of like you know, it's this this you know, amazing ray of hope that um uh yeah, that change can happen like this.

Hope And Politics In Budapest

Roopinder 2:21

Yeah, yeah. I think Orban was probably in the lead of this populist movement that seemed to sweep the world and ended up landing in the US. It was uh uh definitely a big rebuke of um we are in an era of strong men. That's for sure. That's for sure. All right, so we so Ron, you published the reason we're talking today is you you went and published a I would say very, very exhaustive study on marketing of the age of AI. I've got it here. I don't know if you can see it. See the I have to apologize for the coffee stain on it, but that just proves I've been reading it.

Ron Close 3:03

Oh, that's that's excellent. Oh, that's uh it's battle hardened now. Yeah, I was um so I've been working on helping start it started with my work with the startup, and then just some curiosity as I transitioned out to Hungary about how can we get really good at marketing for AEO? And you know, I I looked around and I found a lot of pieces of information, a lot of theory, but I didn't have a practical step-by-step of what I needed to do. And so I decided to write it. And so what you have the current version of it's around 35 pages. It's growing because every week I'm learning something new. It's you know, it's a pretty rapidly developing and changing area. Um, so I wrote the guide mainly to put it in paper for myself, so I knew what to do, but I thought it would be fun to share and get some feedback from some others. And I've made some interesting connections with other marketers along the way over the past couple of weeks where we've had um some great dialogues about how to do this. And the idea is that, you know, we're in a different world. It's it's uh from a marketing perspective, I think it's a renaissance that takes us back to a new level of authenticity where you are being judged on the acceptance of your message, not just lining up the right keywords so somebody can find you. And so I think this learning about AEO is really learning how to take the message that you develop as a company, how to make sure that it's meaningful, but then structure it in a way so the AI tools can find you. And um, so along that journey, I started to I needed a way to test it and see if we're successful. So I was working on a methodology where I would take the the high-level messaging of either a company or an industry, uh, describe the ideal customers, and then develop the most likely prompts that people would use and test them out across different LLMs. And I mistakenly found some things that I want to talk to you about uh along the way that were unintended consequences. My intention was to understand how the AI tools worked, what kind of content they reacted to, and how I could get well-formulated messages uh into the answer bucket of ChatGPT or Gemini or Perplexity or whichever tool someone might be using. I took two industries, I took just a general MES industry, and then I took a segment of the CAD industry where I excluded aerospace and I excluded automotive and tried to find out if you were looking for a CAD system using these LLMs, what might happen. And that's when um I thought, my gosh, I got to call Roopinder because this is not what I expected. What I expected was to see references from you, to see references from other publications, maybe DEVELOP 3D or engineering.com or industry analyst. And in fact, most of the references were not this. Uh, I found that uh, and we'll stay on the CAD side because we did get some trade publications make their way into the MES study. Uh, not as many as I I suspected. But on the CAD side, there was there were very few. And a lot of the references came from the vendors themselves, or in many cases, chat GPT didn't use any references, it just it just uh used train data. So 94% of the time I wasn't getting uh a reference on chat GPT. Gemini used, I'm sorry, uh the number, it wasn't 94%, it was 58%. And then it was much higher for where Gemini was around 94%, was giving me some sort of source reference. And when I looked at these sources, what I was seeing are sometimes the vendors themselves, uh, but then in other cases, they would actually reference the reseller. So let's say you have a reseller of SOLIDWORKS, uh a GoEngineer or one of these companies, and it seems like AI is picking this up as a neutral reference, which is really yeah, which is ridiculous at my point of view.

Roopinder 7:20

But on the other hand, that could be just sour grapes because hey, they didn't they didn't talk, uh they didn't say I was a source, right?

Ron Close 7:27

So no, no, I and that was my um my aha moment where I needed to talk to you about this because you know, where are the sources that I've worked with personally, with people like yourself and others in the industry, they're not showing up. And then when I found something you and I didn't have a chance to talk about, but I looked a little deeper. One of the sources was cited 10 times out of 50 prompts was CADSoft USA. And I thought this was a reseller and it's not. And this is this site is very interesting, a bit of an enigma. I couldn't find anybody credited with owning it. I think it's run with an advertising model, and even running a whois on GoDaddy, I couldn't figure out who it is.

Roopinder 8:11

Um content on it, right? It's a ton of content on there. That's CAD.

Ron Close 8:16

They've got a ton of content, and all the a majority of it is either how to get a free version of this software or comparative, uh comparing on shape to SolidWorks or something like that, which I think is very useful. And that you know, kind of led me led me down another path is when I saw that the most cited type of content were comparisons, uh, comparing one uh one tool with the other. And so yeah, at least but yeah, so that was kind of the initial findings. And my thought is okay, if if as as a CAD vendor, what is the best approach to aligning my message with the market so that these AI tools can consume them? And I didn't come up with a good answer.

Roopinder 9:01

I think you called it feeding the beast.

Ron Close 9:04

Yeah, exactly. Yeah, how do we feed the beast and do it in authentic in an authentic manner? So that was kind of the main reason I wanted to chat with you about this.

AEO Guide And The Testing Method

Roopinder 9:14

Right. Oh, that's that's yeah, welcome to do it. And by the way, I can only speak to the CAD market segment, not the MES segment. I'm halfway, halfway intelligent on that subject and nowhere, nowhere near intelligent on the other. So we'll stick to CAD then if you don't mind. Yeah, of course. All right, okay, so let me back up a little bit and tell people you're you have an extensive history in in marketing and in CAD. And uh I have a similar experience in CAD and in and in uh publishing. Uh where do we stand with AI? Because AI is taking over a lot of what we do, right? And I think a lot of journalists are wondering this. I think probably, to me, marketers are worried about this. Like AI can do all this research, right? Theoretically, it could take the place of analysts, really, because it is doing a lot of analysis. But initial foray into this, you could really think, hey, I've got a I can do a 50-page report on a subject and not hire an analyst, which I would have to pay tens of thousands of dollars for, right? And initially, when you see the report, when you get that thud factor of that report, a hundred-page report, and it's well cited and it can be uh extremely well researched. Um can have good sources among bad sources. If you look carefully, you'll find out oh, these are questionable sources, right? Yeah, but it takes that second book, initial glance is like, damn, this is great. I need to I need to use this. When I left business to business publishing to start this new venture, our CEO told us, hey, we need to use AI to write articles, right? So in that way, I thought, well, where's the future for journalists? I wasn't worried about myself because I'm always fake trying to be a journalist. So I'm really just an engineer, right? But for everybody else that's starting out in the business, what hope have they?

Ron Close 11:11

Yeah, and boy, that's that's a probably a bigger question than I can answer, but I I still think there's an important place for trade media and analysts because of the that extra level of experience to look beyond uh what someone might publish as a new release and be able to make a judgment that I don't think is as easy to make in an automated manner. I think the um, and I was probably jumping way too fast to a conclusion, but I I think that there's an opportunity for journalists and analysts to actually influence these models with having comparative information and maybe more, a little more structure, and what I mean structure in terms of readability schema files and things like that that make it easy for AI to digest. So a little foundational structure in how you publish, but I think in what's published as as well. And don't take this the wrong way, but I I feel like a lot of the publications are driven by advertising, and I don't often see critical evaluation of I'm totally with you.

Roopinder 12:25

I'm not gonna I'm gonna give you my full-throated approval on that one because that's exactly why I left B2B publishing. It wasn't advertising motivated, right? It was all all of my coverage, it ended up to be devoted to advertisers and their companies, right? All of it, 100%. I couldn't cover, for example, I could not cover Shaper 3D because they weren't advertising with us. I was forbidden to cover that company, even if they uh Ishkvan could flew me there and put me up there and was all paid for, right? It wouldn't cost a single dime for to have that coverage. I've had to pay some expenses probably. But point is, it wasn't for a lack of, it was just because of the lack of advertising. They were not advertisers, they couldn't be covered. So that's industry is totally, in my eyes, been corrupted by advertising.

Missing Citations And Biased Sources

Ron Close 13:19

Yeah, and I think we're we're in uh an interesting time now. If the LLMs are not seeing these publications as credible, and more and more companies and engineers, designers are going to do their own research using LLMs. Um it's you know, me, if I were working at a CAD vendor, I might be very cautious about investing advertising dollars. Um they're not they're no longer recognized as the authority that they once were. Um I'm gonna age myself, but I remember back in the Unigraphics version 10 days and Stephen Wolf had this this uh report called the CAD report. And that scared the heck out of us, man. It was like because if if you're he called it like it was, yeah. You know, if if you know you're trying to get something by him, you you you didn't stand a very good chance.

Roopinder 14:19

It was actually my inspiration when I when I got into publishing uh quite accidentally, and that's another story. They I he was my idol. I wish I could have that kind of influence or that kind of knowledge that that Stephen Wolf had at CAD reports, and uh, I never got that, right? I never got that to that because of that advertising orientation or influence. Uh, but but yeah, he was he was great. Uh Dave Weissberg was another one who was yeah, yeah, absolutely. Very influential, uh, wrote considerably, had a great mind for details. Yeah, he was an authority. Those are the old days of Daratech, remember? Daratech itself was an authority on the subject.

Ron Close 15:04

It was a big deal to be um to be mentioned by Daratech uh so would you say the game was a lot easier then?

Roopinder 15:11

I mean, you you played that game, you know, you bought advertising from companies, and it uh you don't have to name the publications, but there was this game you played, you as an everyone played, where you bought advertising and in return the uh publishers would review your product, they would get to know you, right? And and that I would say that once you learned that game, you could play it quite well, and people know how to do it nowadays. What do you do? Right now, if if people aren't looking at publications, they may be Googling, right? But now more than ever, they're using ChatGPT.

Ron Close 15:49

And if you're using Google, you're using Gemini. Um you've got this whole uh zero-click world now, right? So, you know, when you go to Google search, even what before AI, and I'm gonna misquote the number, but I think it was only 60% or less. And now that when you get an answer on AI, there's um it's it's gone down significantly. And and so when you're Googling, yeah, you're you're participating in with Gemini, you get that AI preview, and it's very uncommon for people to click through, they're getting the answers that they need without without going to the website of uh of whoever's uh the source for that information.

Roopinder 16:34

Yeah, so you and I, experts, made chafe against this whole thing that people give it a perfunctory look at it, and it's good enough. It reminds me of the days when uh when bloggers and uh everybody everybody was a news person, and then and then you you trusted what you got on your Twitter feed, you know, and more or Facebook, and suddenly nobody cared about the good publications, the reputable publications like the Times or the Wall Street Journal, the Washington Post, right, Los Angeles Times, because you had to pay for that. But here you're getting tons and tons of information for free, and it wasn't great. You kind of knew it wasn't great, right? But it was good enough, right? It was good enough, and it was free. So this is like where we're, I think this is similar to what we're facing now with chat GPDs. It's it's free, unless you and I pay pay for our versions, but uh, you know, for the most part, free answers, right? And okay, maybe not be good enough. It might be like, hey, okay, this you know, BHS is not as good as Betamax, but you know, but it's the standard, right? Everybody's using it. So why not? Why not? But you ever see that movie Idiocracy? Oh, yeah. A little closer home, that one. I'm not gonna say we're there or approaching there yet, just because I don't want to be too controversial.

Advertising Corrupts Trust In Media

Ron Close 18:03

I also think that the users of LLMs, which are most of us, we're all getting smarter about it as well. And yeah, and you know, I think people are checking source. Uh, wait a minute, I just saw this comparison, but the vendor was the one that published the comparison. Are you sure, Ron? I yeah, I know you are, I know I am. I think I don't I I think the CAD buyer is not the common person. Okay, okay, okay, and so I I think if they're not there yet, I think they will be. And there's still a window of opportunity to influence them. For yeah, I'm advocating for you and the analyst and journalist community to be that trusted resource. I think there's still uh a window of opportunity to do that. And you know, I go back to CADSoft USA, I think it was a fabulous idea. They're they're putting together content, answering questions that people have asked. And I don't think it replaces what uh we can get from a good experienced person in the industry. I think that with another layer of validation and perspective on top of that research, you could really have something um have something dynamite. You know, I don't think it replaces, but I think it definitely empowers the industry to to do more than it has.

Roopinder 19:28

Okay, now Ron, you you wrote these reports because because you you had some questions and you couldn't find anybody that had done this kind of research. So you did your own. I did, yeah. Yeah. And uh this is uh now, did you coin this phrase AEO answer engine optimization?

Ron Close 19:46

Or is that I wish I had, man. You know how hard it is to buy an AEO-related uh domain now?

Roopinder 19:53

Oh, at first I was like, hey, why aren't they why isn't why aren't chatbots finding stuff that I wrote? I I'm an expert. Why isn't it finding stuff I report? But then I I realized, oh, well, it doesn't really trust us. Uh I I don't think I trust a lot of publications. I I don't. There's very few objective analysts out there uh that I that I can trust. Um I have to say, one of the faults of it is that it's pretty much agnostic about where it's getting its information to the except of course it excludes publications, right? Otherwise, it gets it gets answers from every damn place. And the fact that it gets answers from sources that I would automatically not use, right? For as for example, you ask it what's a good CAD program, it can very well tell you answers that it got from a CAD program company, right? If that company has, and you found this to be true, companies that I do comparative reviews on their site, right? Which are to us, to us in the business, it's like, okay, that's already you can just not read that because it's gonna be biased, right? It's gonna have exclusions, it's not it's not reliable. You can't have everybody, you can't have a parent saying their baby is pretty and believe them, right? Yeah, everybody does that. So that's already discounted because of a source. Chatbots are not doing that. They're like, hey, that's good information. We're gonna present it, and and maybe they'll cite it.

Ron Close 21:33

Claude does citations quite frequently, Gemini does because of the different way the models are trained. I think it's more because there's a vacuum of information than because they are just looking for whatever they can find. I mean, if you look at, I don't know much about how this works on the consumer side, but I did have a chat with uh the guy who runs a business that is helping companies become more. Highly recognized by involving himself in Reddit conversations. And those conversations are helping them to quickly gain traction in a place that would otherwise be very difficult to do from their own websites. So it could have to do with the maturity of the information out there, the maturity of the industry. But I do think, and one of the things that I'm working on with the company I'm working with is there's a lot you can do as a publisher, and I don't mean you, but I mean any of us that are publishing websites or whatever to help to give authenticity signals. You know, for example, if you you you're stating some facts, uh uh just a line on there that these facts are verified by this company on this date, or you know, something that says this is verified information that gives you uh more of a of a I don't won't say a right, but more of a notation as an authority on there. I still I I think there's a lot of room to do that still.

Roopinder 23:09

Um gave me a great idea. What I'm gonna do is I'm gonna, when I do some writing, I'm gonna cite Claude and I'm gonna cite Gemini, right? And that will create create this loop where we're recommending each other, right? You know, because if you you know if you buy Google, if you buy Google ads, you're much more likely to be in there. That's a suspicion, maybe a conspiracy. If you buy Google ads, you're more likely to be featured in their in their regular list. Uh they say that because you're feeding, you know, you're feeding the beast, as it were, right? You're giving them some profit. So they have a right, they have an uh inclination to favor you. So okay, I'm not okay. Maybe, maybe. I agree with you on that uh subject of there was a vacuum of information, right? Business to business publishers, because of their advertisers, again, back I'm gonna blame advertising for everything, they were not doing comparative reviews.

Ron Close 24:07

They were absolutely and that's you know, that's AEO 101 is is write comparative reviews. And you know, in the past, I would have been very cautious about mentioning a competitor in any of my marketing materials, but that's exactly what gets you mentioned, and um, because nobody's doing it, and that's what the kind of questions people are asking is give me a comparison of SolidWorks versus Onshape versus ShapR and tell me which one, you know, which one works. And if you are someone is answering those questions, then it won't have to go to one of the vendor sites as the main authority.

Comparative Reviews As AEO Fuel

Roopinder 24:48

When I first started on publishing, like I said, accidentally, never wanted to be editor-in-chief, but I became editor-in-chief of Cadence magazine. One of the first things I wrote about was a comparative review. This is all inside information, not been published before. So you got a scoop here for what it's worth. We did a comparative review on SolidWorks. SolidWorks was just coming out. I did a comparative review pitching mechanical desktop and AutoCAD against SolidWorks because it's just coming out. Yeah. And it was a kind of a, how should I say, a foray into other things because all we're just all we're covering was we were an Autodesk-based publication, meaning Autodesk was a major publisher, we were focusing on them, but SolidWorks was just coming into the world, and we thought, hey, that our publisher thought, hey, there could be some business there. Let's do a review, let's check them out. So there I was. I was like, you know, the intrepid journalist. I was going to compare the two products, and I got into so much trouble, Ron, because because okay, Autodesk hated the fact that we were considering any other software. Yeah. They hated that. SolidWorks hated it because we compared them not so favorably to AutoCAD. I you probably can't believe this. And looking back, and I can't believe it either. But we were looking at it from an AutoCAD user's point of view. Like for AutoCAD, it's going to be much easier to transition a mechanical desktop. It's built on it, right? Rather than learn a whole new CAD product. So that's what we're going in with that premise that if you're an AutoCAD user, go into that that was not understood by John Hirschtick, right? John Hirschtick, we met, I met him for the first time at a NDES. You remember NDES? Oh, yeah, yeah. And it was, I knew it wasn't gonna go down well, right? But the publisher said, Here's somebody I want you to meet, and here's Hirsctick. And Hirschtick is a tall figure. I'm not, right? I I'm like, look, oh my god, how much trouble am I in? Hirsh, right? Because Hirschtick didn't like that. By the way, I had this really just as an even more awkward meeting with Dominic Galelo, who from Autodesk. I don't know if you know him. I know of him, yeah. You know of him? I think he was there when I was at Autodesk. Oh, we had a very acrimonious relationship. At the time, we're we're like buddies now, well, not buddies, but he didn't like that that we were fake. We were I kept asking him, like, why doesn't mechanical desktop have lofting, right? And he I kept bugging him, like, you need to put this shit in because uh SolidWorks already has it, right? I was doing comparative remote, and I was pressuring him to get up to speed because those guys are gonna eat your lunch, right? And he hated me too because I put this in in print, right? So this is I'm not making any friends, right? Not making any friends. Eventually I had to pretty much had to was pressured out of Cadence magazine because I was not no one was happy with me. But anyhow, point is it's really difficult to do the kind of stuff you should be doing for users.

Ron Close 28:08

I don't know how I mean I you know, I'm scratching my head as to how you you can make a business out of this because none of the vendors are gonna want to pay you anymore. And then um, of course, nobody wants to pay for the information on the consumer side. Yeah, yeah. I I wish I if I had that answer, I'd boy, that would be amazing.

Roopinder 28:28

What I'm banking on is that everybody except the leader will want reviews of their products, right? And comparative reviews. So yeah, it would behoove everybody who's not the market, not a market leader, nTOP, for example, or Shaper 3D, right? So that they would want to be compared to because they need to be. Now they're putting their stuff, comparative reviews, on their own site. Eventually, I'm convinced the chat bots will realize, hey, that's pretty dumb, right? To have comparative reviews on the site, right? Let's get Roopinder to publish it instead, right? And then it'll actually be credible because it won't it'd be Repender saying your baby's pretty. If their baby's pretty. No, I'll have to keep my objective hat on and say, you know, exactly a good judge of beauty there. But really, that that needs to happen. There has been no third-party, credible third-party source of information for the CAD user. So it's created this dearth of information, right? It just doesn't matter. Exactly.

Adjacent Tools And Startup White Space

Ron Close 29:32

Yeah, exactly. Um, what you know, and I'm sure we could do a lot more thorough research than I did, but I think if we dig deeper, we're going to keep finding that for the CAD industry. One thing I um it's a little bit of a tangent, but I want to take you there because I think it's important, is what I didn't expect was the number of adjacent products that showed up in uh in product mentions. So if I was I was key, I was structuring my prompts so that they were focused around CAD tools and design process specifically, but a lot of tools for analysis or visualization and even PDM uh floated, and I think I came up with close to 200 total. Really? Yeah, uh quite a large number of mentions for these other tools. And I do think there's a really interesting opportunity, perhaps not for CAD design, but if you're a vendor, and I probably shouldn't mention any, but let's say you're doing AI-based design or you're doing finite element modeling, or you're creating visualization on CAD models, I think there's a huge opportunity right now for these companies to be mentioned in the LLMs associated with CAD. As what the LLMs are trying to do is they're trying to, even though I'm asking a narrow question, it's widen out. So, well, if you're thinking about CAD, you should also have these other capabilities. Um, because I asked specific questions about manufacturing, about uh specific questions about the consumer electronics industry, uh, industrial machinery. And it said, well, if you're if you're in industrial machinery, these are the data management solutions that you should be looking at. So I think there's, and I don't know how that translates to you as a publisher, but I I I think that to me that was super interesting. That looked the most authentic of the information that I found, is it's it's connecting the pieces between what tools work together.

Roopinder 31:42

Yeah, that's very, I think that could be a very could very well be how LLMs are used to help engineers because it's not all about CAD, it's not all about PLM. There's all these auxiliary tasks and functions that we need to do that are related. Uh almost like simulation is an auxiliary test for designers.

Ron Close 32:02

Yeah, yeah, and maybe, maybe that's the um opportunity is to because CAD's so static as well, right? You have the same five or six products for the last 10 or 15 years with very few exceptions in Shapr 3D being one of them. But the changes in the ecosystem, maybe that's the battleground, and maybe that's the place that is gonna be more interesting for uh both from a publishing perspective and from a marketing perspective for these companies. If I'm marketing engineering analysis and I can associate myself with the design process, I'm I'm gonna get I'm gonna get mentioned straight away by these LLMs.

Roopinder 32:42

Yeah. Another failure, as it were, of the business-to-business publishing is that they won't look at all the startups that are doing wonderful things or really uh timely things with AI in all of this, you know, this this companies have sprouted with AI that are doing auxiliary tasks or uh or uh even uh related directly related tasks to to CAD, uh PLM and simulation. One person is doing it outside of publishing, you know him, is uh Fino, right? Yeah. Michael Finocchiaro. Is that how you say it? I I think that is closer than I will get. He's amassed a list of companies of 200 subcompanies. And I thought, oh my god, that's something publishers ought to be doing. No, still I get back, what bothers me is like we we look at all these reports, all these products that we to buy from washing machines to cars, right? And we have consumer reports to do that, right? Yeah, and or some are in other areas of tech, even we have ours technical or we have the verge or something. We have good sources that do hard-hitting, comparative and critical reviews, right? Not just like, oh, how great, how pretty is this baby. Like how you know, they do they do critical reviews, they're not afraid to bash a product if it needs it, right? If it's really does it does something stupid. They'll say that. They'll say that, maybe nicer than that, but but my point is he's so fino has done this on his own, right?

Ron Close 34:27

No, it's a really good point. And you know, Fino, if you're listening, I think you have a terrific opportunity to be the voice of reason here and to be a source for these LLMs, because you're right, he's he's talking to these people that are shaping the future. And uh, you know, it doesn't mean you have to build a CAD to all of the surrounding ecosystem applications. Like you said, he's got a list of 200 of them. And uh yeah, I think that's that's white space, right?

Roopinder 34:56

That's already a service because what else do we have? We don't have we don't have endas anymore. We don't have shows where we can look at them all and say, hey, look, here's all the here's all the in our days. It was like, hey, here's all these MCAD products, right? Let me see which one is better. There's nothing like that now, right? So there's just people at Fino and you know, I'm trying, but he's done a really good job of that. I I commended him for it, but that's again, that's another failure of our publishing world, right? I think so. We we need that. We've got it, we've got to do this research. There has to be somebody else out there. We can't depend on uh chatbots to do it because one day they're gonna figure out, hey, we're not going to use render written, render written uh comparative reviews, that's not good, or we're gonna realize that, oh, these people that read it may not be 100% reliable for for absolutely, yeah.

How To Influence Models And Wrap

Ron Close 35:56

And I that's why I think there's uh probably as much of an opportunity as anything to to I don't say turn this around, but to influence them while while you know why why it's still learning. Yeah, yeah. You know, think that Chat GT and Gemini are just two years old right now, so right, yeah.

Roopinder 36:13

They are they are babies still, they're learning fast. Reminds me of another movie who must have seen it, 2001, right? Where at the end we just realized there's a higher being, and it's it's Hal. It's Hal in that case, like the earliest implementation of AI I could think of at the moment. But Hal is doing this thing, and and Hal learns, but there's even a higher force than that occurs at the end. Okay, we'll get it.

Ron Close 36:39

Hal's wife, yeah.

Roopinder 36:41

Hal's wife, this is how. Yeah, oh boy. All right, so yeah, I don't want to take up all your time. So you need to get to your your beer and your dinner and your wife, right? All right, excellent. Ron, great catching up. Thanks for being on the show. All right, and hope we get it. Thanks for sender. Appreciate it. All right, all right, Ron, take it easy. Thanks for listening to ENGtechnica TV. If you'd like to tell your story on this podcast, contact me at roopinder at engtechnica or message me on LinkedIn.