Podcast Conversation Transcript
SUMMARY KEYWORDS
ai, creating, software, companies, open source, chat, ventures, tools, integrating, gpt, subscribe, podcast, growth, tested, putting, jobs, ted, john paul, articles
John Paul Mains 00:18
Okay. Hey, everybody. Welcome back to the growth ventures. Podcast. We got John Paul here today with Ted, we are going to be continuing on our conversations about AI. Yep. If you’re like some of us out here, you’re getting sick of hearing about AI. But guess what, it’s not going away. It’s only going to get worse. Or worse. Or better, depending on your point of view. Yeah, I’ve heard plenty of people that are like, I hate AI. And I never want to see it again. And I saw it on the new Marvel show that came out recently, this secret wars or something like that the intro? video was all AI and all I heard also was everybody. “Oh, I hate this. I hate AI is terrible”.
Ted Jones 01:02
I actually haven’t seen that yet. No, no. Look, I have to look for that.
John Paul Mains 01:07
Yeah, it’s not too bad. But in any case, there was an announcement today than it was today or yesterday. Oh, well, as of the recording of this, there was an announcement about llama to llama to is a new are not necessarily new, an engine that just got put out new Jeep GPT engine from a collaboration between Microsoft and Facebook. It sounds a little dangerous. Not sure I would necessarily trust that one. But the cool thing about it is that it’s open source. And by being open source, that means that you could literally supposedly I haven’t done it, but you could download the thing and actually use it behind your own well on your own servers on your own with your own own code, own content on everything. And free to use commercial research. Definitely.
Ted Jones 02:05
I think open source is probably a key to look out for I think, you know, open AI chat, GBT started is open source. Right? It did. They made the switch over to be for-profit. And sure. Yeah, I mean, from an AI language model, you’d hope that it would be open source and something that you can build off.
John Paul Mains 02:26
Right, exactly. No, obviously, there’s been a huge number of plugins and enhancements and things like that AI open AI is version of it. But I can definitely see scenarios on wanting to have a open source one that I myself my own business could use for doing whatever else. There’s all sorts of opportunities there for integrating it in with my own software platform for doing doing X, Y and Z. There’s all I mean, it’s just kind of unlimited.
Ted Jones 02:57
Oh, yeah. Especially when you’re looking at your own data sets or your own constraints. I know even chat, GBT just recently announced it as a beta, right? They announced their beta features to their paid subscribers, but the custom instructions, so they put this constraint around, right, so that you know, you can any question you ask chat, GBT can be within context of some type of instruction. You give chat GBT, right. Good example is you being a developer, you develop in a specific type of code. A query, you go and ask chat GBT a question. It’s going to make sure it retains the instruction you gave it that you’re a jQuery.
John Paul Mains 03:56
So that’s exactly that’s, that’s the beauty of AI. Well, supposedly, the beauty of AI if you train it the way you want to train it, it will behave the way you want it to behave or communicate or create the way you want it to create. It. Yeah, I
Ted Jones 04:14
just don’t think they’ve ever built that memory into it. Right. So having that past, you know, I think the more chat GBT and continue within a conversation it’ll reference your previous but to have a memory across different chats Is that gonna change how those affect their usage?
John Paul Mains 04:36
Through through? Yeah, you really have to kind of wonder though, I mean, what? Yeah, it seems like with with with open AI and chat GP chat GPT when it came out now we’re seeing, like, literally hundreds of companies seem to be popping up every single day, based in Canvas. I mean, I don’t know where in the world that came from. Are they building on top of GPT? Or do they have their own thing? in the works, I can’t imagine that they did just but all these companies are popping up just like oh, yeah, here’s this great new thing. Here’s this software for cropping podcasts into multiple parts so that you can shoot us short snippets and shorts and things like that on all these different platforms.
Ted Jones 05:18
Yeah, sure. Even apples throwing their hat into the ring with, you know, their own language model. Yeah. You know, that would have to train it from somewhere, right. So I know that’s coming from everything we’re putting out there.
John Paul Mains 05:32
So if open AI really wasn’t open, it’s more like closed AI. And all of these companies popped up using this, imagine what we’re about to see, with a company like Facebook and Microsoft, which they’ve created, I assume a pretty good product. You imagine all the companies out there, they’re gonna basically take that and say, Oh, now we have our own proprietary system, I don’t know, that you can use for doing XY and Z.
Ted Jones 05:59
white labeling AI? Just, yeah, you don’t know where the source comes from. But you’re offering a toolset that has this massive amount of language model behind it. Right. And I think, like to your point, that’s what a lot of these software applications are doing. You know, we’ve got Wix even grammerly, you know, which has been around for a long time, has a very natural language input now. And you just say, Hey, this is what I’m looking for. You know, it’s gone. Through probably API’s, maybe back to chatGPT could be back out what you’re looking for. Yeah.
John Paul Mains 06:39
But you point like, like, yeah, I love grammar, like, yeah, absolutely love Grammarly. It’s, you know, I can’t write worth anything. And it’s, it’s like, it’s conferencing, oh, you got this wrong, got that wrong, changes, changes, it’s so easy. And I think that’s the thing of where we’re kind of going with all this for from a business perspective, at least, if we look at jobs that we do create, a lot of us are very creative jobs, I’m writing something, video creating a video, I’m creating a graphic, I’m doing some form of communications, or I’m using a spreadsheet, creating code. If you’re creating something in digital form, then it’s I think that you know, the generative AI is going to impact every single one of those jobs, it’s become an assistance, every single software platform is going to have an ethic. That’s why part of why Apple stepped into it, because they’ve got all these different software products that you use for different things. They have to provide the functionality because everybody is it makes life a lot easier. Yeah. Many ways. These are assistance in some ways. Yeah. It’s, I mean, they’re cranking out, I find it, I find it personally extremely easily. Yeah, you know, what I’m capable of using my within my job role for a lot of the different things that I do. So it makes my life a lot easier and getting things done a lot more efficiently. Is it perfect? No. But we’ve seen I mean, kind of the classic of fingers issue that image a eyes have like, they they got it wrong for years and years and years. And all of a sudden now they’re finally Yeah, they’ve got it, they figured it out. Yeah, you’re only gonna get better. Yeah.
Ted Jones 08:27
Well, if I think you know, as more companies continue to build this functionality into their toolset and their offerings, you know, what are the what if you’re not, right? What if that company that you’ve increased your business and do just hasn’t decided to go down that, that AI route and you’re left with, you know, continuing to click 10 times to get something done versus, Hey, I just go do it.
John Paul Mains 08:58
Now, if we thought things were fast in the internet age, what’s gonna happen here in the AI age, this the speed of development, the speed of progress that’s going on? I mean, we’re talking massive, massive leap here, truthfully, and we’re gonna, and I don’t think we I don’t think that we’re even close to tapping in the full potential of what this is gonna be capable of doing. is a good thing. Time will tell.
Ted Jones 09:28
Yeah. Yeah. What was it saying software? Software is going to eat the world or something like that. AI is going to eat software, right? Like, AI is just gonna start with a prompt. It’s always whatever you’re going to be creating is always going to be starting with.
John Paul Mains 09:45
I mean, we’ve seen just in the manufacturing world, just the advent of robotics and what it did there. I think that the parallel really comes over to our world where Okay, yeah, robotics really dominated and changed everything in manufacturing space AI is going to really change, it was gonna pack that space too. But it’s going to really impact our space a lot as knowledge creators. for better, for worse, it’s here, it’s not going to go away these big companies investing enormous amount of money in it. And you’re right. I mean, if those companies decide, hey, you know, I’m not going to use it. Okay. It’s choice. You lose efficiency possible. Yeah, you will lose efficiencies compared to competitors. Maybe your quality of work will be better? I don’t know, it depends on how far AI ends up going.
Ted Jones 10:39
So well. And yeah, the one that’s interesting, because, you know, even Visual Studio now is integrating like, co-pilot, right? And so if you have a tested piece of code that bug-free, that does a certain function, sure, why not keep spitting that back out so that, you know, it doesn’t have to be tested again, or it doesn’t have to be rewritten from scratch. So it’d be interesting to see how many of these building blocks from AI are just continually reused and recycled. You know. And I think even we talked about this last time content, even, you know, content, and now Google’s going to be approaching the indexing side of things when there are so many articles out there right now about using AI. Now, it’s just a regurgitation of the same case.
John Paul Mains 11:38
Now, it’d be interesting that, you know, I’ve got to go this way. I can’t imagine not going this way where you’re getting AI, you know, I don’t think I have a hard time believing that we’ll ever get to the true AI that everybody think it was like a terminator kind of world scenario. But can it? Can it feel like it he probably. But I would think that these AIs will get good enough where they can start creating are capable of creating, not just from prompt, but truly, truly able to create brand new articles or go dig in create something new out of research, or things long as I think Dave is going to come and which will be a new level of leap forward. But yeah, so I don’t know.
Ted Jones 12:33
I think after researches I’m not sure if you heard anything about it, but like Google possibly writing articles and presenting those two new sources.
John Paul Mains 12:47
The headline I don’t know what will that will I’ve got to go read that one.
Ted Jones 12:51
Well, yeah, I think I just saw like an hour ago, so it just made why,
John Paul Mains 12:56
why I’m not too sure. But yeah, somebody at some techies somewhere playing with tools and saying, Hey, look how cool this is. Yeah, see what I can do. doing right now?
Ted Jones 13:08
We’re gonna start putting this stuff in the hands of techies, right. Like we’ve got the language model now. People are starting to build. They’re starting to say, Hey, this is what I can do with it. Look out.
John Paul Mains 13:19
Yeah, that’s kind of thing, right? Everybody’s like, Ooh, look at this. This is cool. Hey, maybe I can get money for this. There’s not a whole lot of thought going on. On Oh, should I do that? It’s more like Hey, cool. I did this. Look at me.
Ted Jones 13:34
Yeah, well, that’s what I like getting down to those small basic building blocks of practical use.
John Paul Mains 13:44
Huge. Yeah, he is
Ted Jones 13:45
Grammarly everyday like you said, you love Grammarly, then, you know, start playing with that tool and Grammarly and see how it improves your workflows?
John Paul Mains 13:54
Absolutely. There’s a lot of tools like that. I mean, there’s the production that we have with Ron, this podcast alone, using a lot of different tools to get stuff done. That would take a lot of time. It’s just efficiency. It’s it’s an aid. It’s an assistant on getting stuff done within my own job for getting stuff done faster, so I can move faster. And it works well. If it didn’t work. Well. I wouldn’t use it. But it does. It works great. So I guess Yep. So the moral of story is keep watching this space. But as a business, you got to pay attention here because it’s coming in this can be integrated in with every facet of what have probably every software platform out there that creates or analyzes anything, it’s going to be in it. It’s just a matter of time. It’s coming fast, fast, fast.
Ted Jones 14:45
Too fast to pay attention to everything.
John Paul Mains 14:52
All right, everybody. Be sure to subscribe to us wherever you’re listening to us.