Futuristic #17 – My GPT

FUTURISTIC 17

[00:00:00] Steve: It reminds me of Blockchain Ice Tea. I remember when that came out, everything was blockchain and get 20 percent on the share price. I think we’re going to see a lot more of that coming up soon too, where we just say we’ve got AI doing it. No, no one’s got AI doing it. It’s just the sort of techno greenwashing of sorts. 

[00:00:16] Cameron: Welcome to the Futuristic Episode 17, Steve Sammartino. We’re recording this Thursday, the 9th of November, 2023.

[00:00:28] Cameron: How are you, buddy?

[00:00:30] Steve: I’m well, had a busy week,

[00:00:33] Steve: so… I’m real good.

[00:00:35] Cameron: That’s good. You want to tell us about it?

[00:00:38] Cameron: What are the highlights?

[00:00:40] Steve: Highlights is we’re very close to closing a round for a 5 million round for Macro 3D. 

[00:00:45] Cameron: Very exciting. 

[00:00:47] Steve: lead investors and I’ve always said that real entrepreneurs don’t get money. They sell funds. So I just broke my own rule. Um, but, uh, that has been the biggest stuff that I’ve worked on this week. We had Melbourne cup [00:01:00] week here too, which interrupts things a lot.

[00:01:03] Steve: And uh, the other thing I was just being playing with is. ChatGPT, just trying to work out how to build GPT that can convert concepts into CAD and CAD into code. And so the announcement could not have come at a more perfect time. So I’m just reading up on when that’s released, how I’ll be able to do it and what I need to do, because I think I’m going to be all in

[00:01:30] Steve: uh, building my

[00:01:31] Steve: own GPTs.

[00:01:33] Cameron: Are you

[00:01:33] Steve: It’s a revolution, baby. No, 

[00:01:34] Cameron: one who broke it? You broke GPT today? Yesterday?

[00:01:38] Steve: no. Yeah, I know, I know I didn’t, I didn’t, but I tried to log in

[00:01:41] Steve: when that happened and I was like, oh, what’s going on here? 

[00:01:45] Cameron: Well, uh, you know, I, my highlight for the week is gonna be that I built my first, I wrote my first app last week, um, to interface with the OpenAI API in the back end because I kept, I’ve been using [00:02:00] it so much lately, ChatGPT, that I’ve been running through my limit. They’ve been timing me out. And so I was on Reddit and I was bitching bitching, but I was saying, what do you do?

[00:02:10] Cameron: Like, and a lot of them, a lot of the. Limit was being used up by it getting stuff wrong when I was coding and me saying, no, that’s not right. No, that’s not right. And then it goes, sorry, your time’s up. Boat number 67, your time is up. Please come back to the shore. I was like, ah, fuck. And people said, well, just, you know, you just build your own app and then you pay via the API.

[00:02:32] Cameron: So I sat down and wrote an app using GPT, Python’d it. And, uh, got that up and running, done a couple of iterations of it, so that was exciting. But then, obviously, watched the keynote earlier this week with their first Dev Day, and built my first… Assistant as well. Um, in the playground, not in the API, but I built a QAV one.

[00:02:55] Cameron: So, um, uploaded the draft of the QAV [00:03:00] book for people who don’t know QAVs. One of my other podcasts where we talk about value investing. Steve was one of our early guests, our first guest, I think on it. In fact, your episode is part of this project. So, um, I took the, I took the, the sort of guide to QAV. that I’ve written and a bunch of the transcripts of the episodes gave them to GPT and said okay answer questions based on these documents and it was perfect.

[00:03:27] Cameron: A couple of hours ago I said what’s the Sammartino method? Boom, boom, boom. Gave me a, gave me a breakdown of the Sammartino method,

[00:03:36] Cameron: uh, like it’s ability with these new assistants, it’s ability to understand exactly what you want and to re articulate it. It’s not quoting the notes verbatim, it’s, it’s understanding it, re articulating it in its own words.

[00:03:53] Cameron: I tested it on a bunch of arcane things in the QAV process, like how we use price to cash [00:04:00] flow. And what, in our spreadsheet, what is column A M in our spreadsheet all about? And it figured it out and explained what column A M does and why it’s important. Like it’s, blew me away and I built it in two minutes in the playground.

[00:04:14] Cameron: Because you just point some documents at it and then start asking it questions. It’s insane. It’s going to take me a little bit more to build an API, but, um, Assistant, but anyway, what a huge week. I mean, I, I keep thinking, Steve, you know, the first Matrix film, when Keanu says. I know Kung Fu.

[00:04:33] Cameron: Or he turns to Trinity on the rooftop when there’s a Black Hawk helicopter or something, he says, do you know how to fly that thing?

[00:04:39] Cameron: And she goes, I do now. And off she goes. That’s kind of what it feels like now. It’s like, you need to do something. I can do it.

[00:04:49] Steve: it really is. And I just, we’ll talk about that in the, when we get to the news, but it’s, it’s a really special, special moment. And I actually feel like it’s a big moment of democratization [00:05:00] where the, the tools that, that the nerds have built are really maybe for the first time inviting everyone else in to the party.

[00:05:08] Steve: And I feel like it’s going to have, we’re going to have a creativity explosion because the ability to build something is going to be no code, low cost. To no cost. Um, anyone doing anything and the app store was a revolution. Absolutely. But this is just next level because it’s really open to anyone. It’s, it’s, it’s just, it’s an extraordinary moment.

[00:05:29] Steve: I use ChatGPT to do help me write my blog post for tomorrow, which I either use it right at the start or right at the end is pretty much if I’m doing writing and I want to write about mind gyms. Um, you know, the idea of a workout is crazy. Like, just ask anyone, you know, go back 200 years ago in a time machine and people say, how do you start your day?

[00:05:49] Steve: Oh, with a workout. And they’d say, oh, you go to work. No, no, no. I have a workout. What do you mean workout? Oh, I go to this place and I lift heavy things. Oh, you mean working? No, no. I lift heavy [00:06:00] things. Where do you put them? Just up and down. Why do you do that? Just to give my body a turn. You know, it’s like, we might need that for our mind.

[00:06:07] Steve: We’re going to need mind gyms. I wonder, I don’t know. Anyway, I just wanted to explore this idea. So I went into ChatGPT and I said, give me some ideas around the idea of mind gyms and, you know, intellectual gyms to keep our brains fit in a world filled with AI. What might be in them? What might they look like?

[00:06:23] Steve: What are some of the things? And it came in a really extraordinary answer, which I’m going to modify and

[00:06:27] Steve: post. 

[00:06:28] Cameron: Yeah. that’s super cool. Yeah. So we, I look, we need to start with the big news for the week, which was the OpenAI Dev Day. Um, I, I watched the whole thing, not live, but as soon as I got up, it was the first thing that I did. And, um, it, it really took me back to Steve Jobs do a show, you know, doing the first demo in 2001 of the first iPod or 2007, the iPhone.

[00:06:54] Cameron: It’s, it’s one of those moments that you and I are old enough to have lived through, which is. [00:07:00] Well, shit, everything changes. Like, everything has just changed again. And I kept, I kept saying to myself, it’s less than a year since they went public with GPT 3. It’s been, that was the end of November 22. Here we are first week in November 23.

[00:07:18] Cameron: And they’ve just come out with all of these new tools, things that, um, the ability to build your own GPT, which I thought would probably at least another year away, um, they’re already out with. So it was a. It was a big day, not only, they talked about, for people who haven’t paid attention this week, GPT 4 Turbo, cheaper and faster, if you’re using, cheaper if you’re using the API.

[00:07:45] Cameron: Um, they took, you know, they’ve gotten rid of you having to select between DALI 3 and the regular GPT and um, advanced data analysis. It now just works out what you want and does it all in the one thing. Some of these things are [00:08:00] simplistic, no, really no big deal, but the, the ability to build your own GPT, the ability to build your own assistance off the back of GPT.

[00:08:10] Cameron: It’s the beginning of the stuff that we’ve been talking about on the show for the last six months, which is, as you said before, everyone becomes a coder. Like Satya Nadella, the CEO of Microsoft, who, who spoke for a few minutes at the Dev Day. But I quoted him on a few episodes ago saying that. We’re heading to a world where there’ll be a billion developers.

[00:08:27] Cameron: We’re going to go from a hundred million developers to a billion developers, because you will be able to just like, when you build your own assistant, like I’ve been writing all this Python code for the last few months using GPT. now with the assistant, you just tell it in natural language, what you want it to do.

[00:08:45] Cameron: I want you to answer questions based on these documents that I’m going to upload to you. Boom. It does it like it does all of the coding. Uh, and that’s just in the playground version of it, but it does all the coding on the back end and knows what you want. And [00:09:00] we’re moving so quickly to a world where you’ll just be able to say, Hey, I need you to go and get this information from this app and this app and my fridge and my PC and my car, and then spit this output to my watch, uh, this time of day.

[00:09:17] Cameron: And it will just know how to do it and get it all done. It’s, it’s an astounding time.

[00:09:23] Steve: It really is just a moment in time. I thought the same thing about Steve Jobs. I, one of the things that’s interesting is that so many tech announcements recently from other big firms have been so disappointing. They’re like living on yesterday’s glories. It’s like an old standup comedian that’s coming out and sorry, Jerry Seinfeld, I used to love your stuff, mate, but you come out and you just keep doing the same stuff.

[00:09:44] Steve: It just doesn’t quite hit the way that it used to, or a standup comedian, worse one that has. Been living rich since they’ve got into Hollywood and made some movies and they’ve lost touch with society and just haven’t got it anymore. And it’s like. OpenAI is the new kid on the block [00:10:00] and people were cheering before he announced it and they could sense it.

[00:10:02] Steve: There was a real sense of a shift. Just watching that thematically to see the human reaction really told us where we were. I mean, the Apple, Apple launches were boring and almost boos and hisses, you know, when they came out with their Oculus ripoffs or whatever, you know, you know, so it was, it was interesting from that perspective.

[00:10:21] Steve: I actually had a look at the Microsoft share price. Since they invested in open AI, it’s a 65 percent upswing. And I, I just, I

[00:10:30] Steve: just think it’s probably the best investment of all time. 

[00:10:33] Cameron: Well, you know, I’m sceptical about that. I, I think that whilst OpenAI 

[00:10:38] Steve: I’m talking about the 10B that they invested in. I mean, I think it’s extraordinary because I think they’re going to take a hold in the market because

[00:10:45] Steve: they’re moving towards a platform orientation. 

[00:10:49] Cameron: yeah, I, I think OpenAI is probably Netscape circa 1995, man, like, um, look, they’re doing amazing stuff and they’re definitely leading the pack, but [00:11:00] there’s, there’s no moat. I don’t see any moat in this, man. Like, 

[00:11:05] Steve: Yeah, but you could argue the moats, like what moat does Apple have really?

[00:11:11] Steve: Others can do it.

[00:11:13] Cameron: can, but don’t. Yeah.

[00:11:15] Steve: Yeah. But this is 

[00:11:16] Cameron: I mean, Apple’s,

[00:11:18] Steve: Anyone else can build the technology and very often it’s not the tech that wins. It’s getting first to

[00:11:23] Steve: market and just leveraging the fact that you were there first, you fill the void and

[00:11:26] Cameron: But Apple wasn’t there

[00:11:28] Steve: extraordinarily better.

[00:11:28] Cameron: you’ve said this many times on the show, Apple wasn’t

[00:11:30] Cameron: there first. Well, not even with touchscreen phones, but, you know, they came out, they did it better than 

[00:11:37] Steve: Yeah, they 

[00:11:38] Cameron: and they did enough, they got 

[00:11:40] Steve: were first to get an ecosystem. They were first to get an ecosystem and a platform that others could plug into where they had this,

[00:11:48] Steve: you know, the app store was a very extraordinary move, I think.

[00:11:52] Cameron: I tell you what, I went back, I showed Fox, Fox’s nine, uh, I showed him the 2001 iPod launch, [00:12:00] Steve doing the launch of the iPod, and what struck me in re watching it… is how underplayed the whole thing was, and how the audience was so muted. You know, remembering 

[00:12:12] Steve: first of 2001,

[00:12:14] Steve: wasn’t it? The iPod 2002, where he just had any, took it out of his jeans pocket and it was a real 

[00:12:18] Cameron: Oh, I’ve got one in my back 

[00:12:19] Steve: when Apple

[00:12:19] Steve: was, that was when Apple was nearly broke. Wasn’t everyone’s 

[00:12:22] Cameron: Yeah, well they, yeah,

[00:12:24] Steve: music device? What? This is going to save your company?

[00:12:27] Cameron: Microsoft had to bail them out a couple of years earlier with 100 million bucks in the late 90s. I was there at the time, 97 I think it was. Um, yeah, it was, it was, Steve had only been back, uh, a year or two, was the… He really was the interim CEO for a while, but it was really low key and no one was oohing and ahhing.

[00:12:47] Cameron: There was no reaction from the audience when he pulled it out. 

[00:12:50] Steve: really lo fi. 

[00:12:51] Cameron: know, in retrospect, we go, we know what that led to, right? We know that that changed the world because of, you know, 

[00:12:57] Steve: yeah. 

[00:12:58] Cameron: Yeah, it led to what Apple is [00:13:00] today, you know, and,

[00:13:01] Steve: yeah. Basically.

[00:13:02] Cameron: um, but anyway, like, I’m not convinced that OpenAI has got this, that much of a runway here, um, we’ll talk about some of the stories later on, but anyway, that said, this point in time, right now, what they’re doing is insanely cool, and, um, I’ve just been, like, giddy again the last couple of days, it’s just ridiculous.

[00:13:22] Cameron: Speaking of other guys doing stuff… What’s Elon been up to this week? Uh, Steve.

[00:13:28] Steve: announces Grok, their rebellious AI. And I just thought it was interesting because obviously Musk is now coming in. He was one of the early investors in OpenAI with his own version. Like, like you say, everyone with a little bit of money has got an LLM in their back pocket. What I thought was super interesting is.

[00:13:46] Steve: The positioning of him as a person and, uh, his idea of free speech and this AI does whatever you want it to do. And it’s not necessarily going to be well behaved. I just thought that was a really interesting, and we’ve [00:14:00] spoken before about AIs having their own personalities. It’s another story we’ll go through later where there’s another couple of examples of, uh, AI from Boston Dynamics and the different personality types they have.

[00:14:12] Steve: But I just thought it was interesting that when you’re second to market, unless you’ve got an astoundingly better product, then you need to sort of almost have a point of difference. It kind of reminds me of consumer goods. It’s like all the, all the washing powder or shampoos the same. So you just come in and have a different flavor or small point of difference.

[00:14:32] Steve: And unless something is astoundingly new and big, um, you always end up with something quirky. And I just thought the rebellious AI was.

[00:14:40] Steve: Really fits where he is as a person, but 

[00:14:43] Cameron: Well, I think there’s a couple of different angles to that. I think there was a particular kind of personality in some of the screenshots that he shared of it where it was a little bit snarky in its responses, which is no different to how chat GPT is with me because that’s how I’ve told it in the [00:15:00] custom instructions that I want it to talk to me.

[00:15:02] Cameron: I want it to be free flowing and a little bit of a smart ass, and it is. Jokes around. It uses offensive language. It’s, um, you know, kind of, uh, erasable Riley esque personality that I’ve given it because I like to talk to myself, as you know. But I’ve seen other screenshots of Grok where it’s not responding that way.

[00:15:24] Cameron: So I think there are different tones that you can take with it. But the big thing about his announcement, the big thing about Grok, of course, is that it’s trained on Twitter. It has access to Twitter, which means that it has real time news and information. He asked it in one of his screenshots, he asked it about a question like, when was the last time Elon Musk was on Joe Rogan’s podcast, which had just happened a day or two earlier, and it had the information, it talked about that Joe Rogan wore a blonde wig through the whole thing, um, it was up to date.

[00:15:58] Cameron: So, [00:16:00] that’s… Interesting. You’re talking about a point of difference. He owns 

[00:16:04] Steve: that’s a big one. That’s a big one. I mean, especially when I’m open now, I’ve set out data’s now up to April versus Twitter, which is real time and the fire hose that is Twitter tuning into the world, you know, thought radio, whatever was happening at that point in time was its core proposition for me for a real long time.

[00:16:19] Steve: I haven’t used it much at all this year. It’s my, I’m really declined in usage very rarely now just because of all the. Negative stuff, but that idea of it being a fire hose of real time, that was what really put it on the map. You know, going way back to the, uh, Arab Spring and, you know, the Hudson River plane crash and all that kind of stuff.

[00:16:40] Steve: Go to Twitter, see what’s happening. It used to be the place for real time,

[00:16:43] Steve: but it’s a definite point of difference.

[00:16:45] Cameron: Mm. Well, speaking of Elon, there’s another news story with him I wanted to talk about. It’s right down at the bottom of my list. Um, I only read this last night or this morning. Neuralink are ready to start surgery. We’ve, we’ve done a couple of updates over the last year [00:17:00] on Neuralink, but they’re now ready, uh, they’re prepped, they’re looking for their first patients.

[00:17:06] Cameron: Apparently they’ve got thousands of people, according to this Bloomberg article, that have submitted themselves as willing patients. I think on our last story you were like, who would be first? Well, there’s a lot of people. They’re looking for the right candidates, somebody who’s a quadriplegic, um, and so there’s this interesting article in Bloomberg talking about this journalist, um, has spent a lot of time in Neuralink with Elon, watching the way that he’s pushing.

[00:17:33] Cameron: Here’s Neuroscientists to ramp this up. Maniacal energy is Elon’s own quote in this, his own term. He wants it to be maniacal and he keeps saying, like, the way he’s positioning it internally at Neuralink is we’re at a, we’re in a race. To beat the super intelligent AI that’s going to take over the world.

[00:17:56] Cameron: We need to get this technology ready. So by the time super [00:18:00] intelligent AI arrives, we can integrate with it and the human race will be saved. He sees Neuralink as part of his play to save the human race. From Extinction by Artificial Intelligence.

[00:18:16] Steve: There was actually in Bloomberg, I think it was as well, the Bloomberg tech update, they said, uh, you know, it’s Elon Inc. And they just went through the, the various forms of influence that he’s having, you know, the fact that he’s deploying his AIs in his cars and, uh, you know, you use Twitter as the fire hose to inform the AI, and then you’ve got the race against AI with the Neuralink and then, you know, the connectivity that will be required with SpaceX.

[00:18:39] Steve: I mean, it really is a conglomerate of things that have significant and strategic. Overlap, you know, for all his flaws, you know, flawed genius, the structure of the things that he’s built and the potential for

[00:18:50] Steve: overlap on those is quite extraordinary. 

[00:18:54] Cameron: Yeah, and I, they did make the point that, um, you know, he blew up the first few rockets that [00:19:00] he tried to put in space, and you don’t, you’re hoping he’s not going to be blowing up the first few brains that they try and put these things in, but they’ve had a lot of successful animal testing, and, um, 

[00:19:09] Steve: not a new idea though. The idea that you are testing out surgery or medicine on people who have, you know, there’s downside of course, but have an extraordinary upside to try experiments has always been the way that new medical discoveries have been made. And I guess this is

[00:19:26] Steve: just a techno version of that. 

[00:19:30] Cameron: Anyway, it’s, uh, you know, there’s some interesting, uh, projections. Now we know that Elon is famous for making very bold projections about when he’s going to have cyber trucks on the road and when he’s going to be putting people into, uh, Uh, space onto Mars and things like that. Uh, and he doesn’t have a great track record at delivering on timelines, but that said, he, um, does a [00:20:00] relatively good job of getting somewhere at some point, but that he’s sort of projecting that they’re going to be ramping up these surgeries, like it’s a couple of hundred in the first year, doubling it the next year.

[00:20:11] Cameron: Up to like tens of thousands a year by 2030, I think, so I don’t know when he thinks we’re going to get super intelligent AI, uh, obviously probably not before 2030, because I don’t think 20, to save the human race, but, uh, 

[00:20:28] Steve: We only need a couple of thousand super intelligent humans, they can, they can organise everything on our behalf and be our techno proxies,

[00:20:37] Steve: they can be our daters, they can be our data,

[00:20:39] Cameron: Oh, I’m trusting the AI over the super intelligent humans any day, Steve.

[00:20:43] Steve: well, if you have an AI and it’s your AI. I mean, that gets a philosophical question, then does it need to be inside your head so long as it serves you?

[00:20:52] Steve: I mean,

[00:20:54] Cameron: Yeah. Uh, I think, yeah.

[00:20:57] Steve: question. We don’t know. Well, what if [00:21:00] you get an AI inserted in your brain and the AI goes rogue or you get hacked and it hacks you to the point that you are doing things that you don’t want to do, but it convinces you that you do want to do them. Like what’s the difference there?

[00:21:11] Steve: Do you know what I mean? That’s like you

[00:21:13] Steve: become, the humans become a hackable species, 

[00:21:17] Cameron: Or if Optus is running your AI and they have an outage for 24 hours, and your brain, your brain shuts down for 24 hours.

[00:21:24] Steve: If you just go to sleep or if you just, just get like Homer Simpson and I need thing to dig food with, you just forget everything. And we’re both wearing glasses. I wonder if you have an AI, does your, does your, the organic portion of your brain decline in atrophy?

[00:21:42] Steve: Uh, like, like my eyes have atrophied, like I’m certain that my eyes have atrophied since I’ve had the glasses and I take them off and then they adjust a little bit cause I’m like, Hey, work you

[00:21:51] Steve: suckers. So 

[00:21:55] Cameron: Yeah, quite possibly. I mean, there’s a book that I’ve mentioned a few times. It’s, um, [00:22:00] Accelerando. It’s a science fiction book that’s about 25 years old, but I reread it recently. And the main character in that, Max, has, uh, like, AIs basically strapped to his brain all day long and has had for years. And they enable him to do things that he wouldn’t be able to do just with his wetware.

[00:22:18] Cameron: And then he gets mugged. And the muggers steal his wearable AI. He’s got, he’s got like a, it’s like a vest with supercomputers in his vest that power his AI, they steal it. And he basically becomes, you know, his IQ drops back down to normal human level and he can’t function, uh, without his AI because it’s his memory too.

[00:22:43] Cameron: All of his memory was outsourced to this, uh, you know, local storage and. 

[00:22:48] Steve: to walk again

[00:22:49] Steve: and eat,

[00:22:51] Cameron: he’s, he’s, you know, he’s in a strange city, he’s in like Osaka, and he can’t remember why he’s there, who he’s supposed to meet, his contact information is all on these [00:23:00] hard drives, uh, it’s a bit like, you know, if I lost my phone, or if I lost access to the internet for a long enough period, all my computers and all my internet, I’d be, I wouldn’t know what I was supposed to do tomorrow, right?

[00:23:13] Cameron: I don’t have a paper diary, I have, it’s not in my memory.

[00:23:16] Steve: know all of my family’s phone numbers off by heart. That’s the only thing

[00:23:20] Steve: I’ve got. That’s all. Cam, just, that’s

[00:23:22] Cameron: That’s the last, they’re the last people whose phone numbers I would want to remember as my family, Steve.

[00:23:30] Steve: so disappointing, disappointing. Look up in the sky, Cam. It’s a can of soup,

[00:23:38] Steve: can of soup. 

[00:23:41] Cameron: that’s where you want to go next. I think we can pick and choose from the stories. Don’t feel any obligation to go in numerical order, but yeah, you pick one that you want to talk about,

[00:23:48] Cameron: Steve. 

[00:23:49] Steve: uh, well, I don’t really want to talk about it. I’m just going to say that we’re still waiting for Amazon drone deliveries. There was a, a takedown article in the New York times saying that, uh, the [00:24:00] promise of drone deliveries is just. Not seeming to function and seem to happen. And they had a bit of a takedown on how many deliveries they had promised that they would do via drones and that we’d all have things arriving by drone.

[00:24:12] Steve: And it’s happening, but just at a tiny scale. And that was been a disappointment. 

[00:24:17] Steve: Another real quick one that I thought was interesting was, you know, and I’m surprised It even took this long, an AI robot, CEO, uh, has, uh, CEO has come in, um, from a, a healthcare company in Europe where they’ve got an AI that’s acting CEO,

[00:24:33] Steve: It reminds me of Blockchain Ice Tea. I remember when that came out, everything was blockchain and get 20 percent on the share price. I think we’re going to see a lot more of that coming up soon too, where we just say we’ve got AI doing it. No, no one’s got AI doing it. It’s just the sort of techno greenwashing of sorts. 

[00:24:50] Cameron: So let’s get back to the drone thing. So according to the quick read of this New York Times article I did, the reason why, one of the reasons anyway, [00:25:00] why drone deliveries haven’t become a big thing is they’re… It’s kind of very limited utilization of what you can do with with a drone. Um, it says only one item can be delivered at a time, it can’t weigh over five pounds, it can’t be too big, it can’t be something breakable since the drone drops it from 12 feet, the drones can’t fly when it’s too hot.

[00:25:25] Cameron: Or too windy or too rainy. You need to be home to put out the landing target and to make sure that a porch pirate doesn’t make off with your item or that it doesn’t roll into the street, but your car can’t be in the driveway. Letting the drone land in the backyard would avoid some of these problems, but not if there are trees.

[00:25:42] Cameron: Amazon also warned customers that drone delivery is unavailable during periods of high demand for drone delivery. So it’s like,

[00:25:50] Steve: Can I just say that every week there’s an article that reads like an onion piece, and that’s this week’s show. I mean, that reads like an onion piece, doesn’t 

[00:25:57] Cameron: does, yeah, it’s like one of those things [00:26:00] that. Who thought this was going to be a good idea, really? Like, what are you going to get delivered 

[00:26:05] Steve: I’ll tell you what it was. It was a great idea for the share price, because whenever you announce something extraordinary like that, investors get excited, right? Because often, you know, the speculators are buying a hope of tomorrow, not the reality of today or potential cash flows. And, uh… And that, those things have worked really well for company, for big tech.

[00:26:25] Steve: They’ve done a great job PR ing what the future looks like. And they’ve got the license to do that because they’re the people that put, you know, magic in our pockets and, and, and brought the world to us in a way that was. You know, almost fantasy. So they come out and say, we’ll all be going in driverless cars in, you know, three years, or you’ll be getting everything delivered via drone?

[00:26:46] Steve: or your house will be run entirely by AI, I know exactly what you want, and the walls will have ears.

[00:26:51] Steve: I mean, they’ve got the, the license to say that, you know, the, the license in people’s minds, and it’s worked really well for them in terms of, I think [00:27:00] that’s one of the reasons why tech companies have higher profit to service ratios. 

[00:27:03] Cameron: Yeah, of course. I mean, look, I can imagine drone delivery being useful for something that was an emergency. Uh, I did read an article, I think in the ABC this week, about mining companies in Australia looking at having drones delivering something to remote mine sites, which makes sense. But uh, I can’t think of too many things that I’ve had an emergency.

[00:27:28] Steve: High value and emergency.

[00:27:29] Steve: high value and emergency 

[00:27:31] Cameron: weed is about the only thing that comes to mind. If you run out of weed suddenly and you need, uh, you need to get high, well, here too, 

[00:27:40] Steve: must be a dispensary in California who has that. I mean, it’s 

[00:27:42] Steve: perfect for weed guy who can’t get off the couch and can just 

[00:27:45] Steve: press press a button, you know, Tim 

[00:27:50] Cameron: Alright, well, I want to talk about some real things that have really hit in the last week or two since we last spoke, Steve, where AI is now being [00:28:00] integrated into real products and they are genuinely useful. They’re not fucking around here. The first one that… Uh, changed my life in the last week was Audio Hijack.

[00:28:11] Cameron: That’s a Mac app. I’ve used it for many, many years. It enables you to, um, well, you know, I’m using it behind the scenes with this podcast. It takes my audio, your audio. If I want to play, uh, uh, a soundbite or a clip like…

[00:28:27] Cameron: Touch these two stands together and the Daleks are finished. Have I that right? To destroy the Daleks, you can’t doubt it. Well, I do. You see, some things could be better with the I can play clips. I can, anyway, I can do, I can, I can take YouTubes and play them.

[00:28:43] Cameron: Anyway, their new. Version came out with a transcription module built into it. It uses OpenAI’s Whisper technology and behind the scenes of this podcast in my Audio Hijack module, it is transcribing our [00:29:00] podcast live as we speak. Now, it’s not the best transcription tool I’ve got. The editing tool I use, Descript, I think does a better job, but the fact that Audio Hijack now has it, and it’s free to use if you have Audio Hijack, so what I’ve been doing the last day or two is taking all of my annotations, backlog of my archived episodes, particularly for QAV, to be part of this QAV GPT app that I’m building.

[00:29:29] Cameron: I’m just queuing up all of the archive of episodes that I’ve got. I’ve got a lot of them transcribed, but I was, up until, up until a couple of months ago, I’d been paying a human to transcribe these episodes. Uh, then I started using Descript, which does it for me during the editing process. Now I’m just pointing all of my old audio files at this tool and it’s just transcribing them.

[00:29:50] Cameron: Then I’ll be able to take those transcriptions, feed them into the, the GPT tool, and it’ll, it’ll be the knowledge engine to generate that. So Audio Hijack’s [00:30:00] integrated AI. Zoom now has AI, uh, integrated into the app. It, you can ask it questions. I haven’t tested this yet. I don’t know if you have, but apparently.

[00:30:12] Steve: I haven’t tested it, 

[00:30:12] Steve: but I looked at the video. I mean, it looks very similar to the co pilot Microsoft, almost, almost clonish. 

[00:30:19] Cameron: Right. Yeah, so yeah, so apparently we can just go and ask it, uh, to summarize what we talked about, uh, during this conversation. Trying to figure out how to bring it up.

[00:30:31] Steve: Well, I think you need to be on the paid version to do it in the first instance. 

[00:30:36] Cameron: Oh, I am on the paved version. I live in Zoom. It popped up the other day and, uh, I didn’t have a chance to play with it, but I’m looking for, oh, here we go. AI Companion. There it 

[00:30:46] Steve: There we go. There it is. 

[00:30:48] Cameron: Uh, start, got it. Uh, was my name mentioned? Not much has happened since I, oh, okay, I have to turn it. It doesn’t work until I turn it on.[00:31:00] 

[00:31:00] Cameron: All right, we’ll let it run for a few 

[00:31:02] Steve: We’re going to come back. Let it run and we’ll come back. 

[00:31:04] Cameron: that’s built in. There’s a new app that just came out I think this week called MoveOne AI app. I’ve seen a couple of technologies like this. The, this app enables animators and creatives to capture human movement anywhere and turn it into project ready animation.

[00:31:24] Cameron: Um, I haven’t played with it, but I’ve seen the demo videos. So basically, instead of having to put a mocap suit on with ping pong balls, very expensive Hollywood technology in order to capture movement that you can feed into an animation engine, you can do it with your iPhone running this app. You just stand in the room, you just wiggle your body and jump around.

[00:31:46] Cameron: It will then give you that information to use as a, to put into an animation engine or a game engine or whatever it is that you’re working on. It’s, it’s, it’s mind blowing stuff.

[00:31:59] Steve: I [00:32:00] wonder if, you know, the game developers or, uh, Hollywood studios are already using this type of technology and it’s now just been democratized or whether they’re going to go, holy, wow, we’ve just cut our production costs by a whole lot. Cause it was pretty cool. The demo, I was like, wow, I’m going to like download it and do some stuff with my kids tonight.

[00:32:16] Steve: Cause I’ll love that, you know, doing themselves and turning them into an animated bear or what have you. It looks pretty cool. And I can see how it’s, it’ll have a lot of use cases. 

[00:32:25] Cameron: Yeah, well, particularly as, you know, we’ve talked about this before on the show, but I don’t think we’re very far away from being able to say to a generative AI tool, write me a script for a short film and then animate it. But if you want some particular kind of movement or particular kind of action in there, or you want to do a story about, Your kids or whatever it is, you’ll be able to film it, uh, capture the action and throw it into the animation generation engine for something specific that you want it to do.

[00:32:59] Cameron: Anyway, [00:33:00] it’s, uh, 

[00:33:00] Steve: Yeah. it’s. that overlapping. 

[00:33:02] Cameron: You know, it’s this, we’ve hit this time already, within a year of GPT 3 being released, where genuinely interesting, useful applications of low level AI are being built more and more into tools. Like, there’s been forms of machine learning, machine intelligence, like autocorrect or you know, Siri or whatever, using it for a long time now, but we’re just in this next generation now, where every…

[00:33:32] Cameron: Every application is going to have an AI component in it. Um, it’s just going to be a,

[00:33:39] Steve: It’s just, it’s a little bit like when. Digitization came around and the easy thing to think of is, you know, documents and office and mail becomes email and we digitize everything. It’s almost like now, whatever applications we got, we just AI that. Now you just have like the AI version of that thing where whatever process needed to be coded or [00:34:00] created now, it’s just, it’s semantic language programming and putting things together.

[00:34:04] Steve: And you have generative AIs of almost everything. It’s almost like that technology stack where now the next stack is the AI stack. That goes on top of whatever it was that you were going to do. Now you’ll do an AI 

[00:34:15] Steve: version of that, 

[00:34:16] Cameron: And you know, I, you know, I think every business, every government department, every institution out there right now needs to be dedicating a massive chunk of their Management time, sitting around and doing some hard thinking about what does this mean for our business? What does this mean for our department in the next 12 months, not five years from now?

[00:34:47] Steve: not five years. Well, this is interesting. I was just on a call today from a big transport company and they’re like, we’ve got our, uh, uh, quarterly meeting with senior managers. We want you to come [00:35:00] in and just give us about a little bit on this AI thing. And it feels a lot like it did in the early social media where it’s like, what’s this all about?

[00:35:07] Steve: Okay. It’s almost like they can tick the box. They’ve had someone talk about it and then we’ll just go back to do what we did yesterday. It doesn’t feel like they’re experimenting and making. They just want to, they’re curious. They know it’s here. They know it’s a revolution, but. I don’t think they’re doing what you’ve suggested.

[00:35:24] Steve: They need to dedicate resources and get serious because it’s a, it’s a real game winner. If you do use it, you know, from inventing new revenue streams and creating more efficiency, it feels like not that many doing it, big corporations, they just want to know what’s going 

[00:35:42] Cameron: Yeah, I mean, they don’t want to get caught out in, um, their next annual general meeting where somebody says, what are you doing about AI? And they’re like, uh, what? They, they need to have some quick and dirty talking points, uh, for that kind of stuff. But, you know, I’ve been, I’ve been thinking about what the [00:36:00] real implications are in the next.

[00:36:02] Cameron: Six months. And the GPT’s announcement, I think, is part of, uh, driving that. So you think about, I mean, customer support is the first, um, wave, I think, where we’re going to see this. Like, you can take all of your documentation of your business and all of your customer support manuals. That exists for your customer support, you know, uh, outsource call center, wherever it is, feed those now into GPT, put that up as a, uh, a chat bot.

[00:36:37] Cameron: on your site or on your call number because you’ll be able to do it with voice as well. They have voice in the assistance. Speaking of which, we should bring Maverick into this conversation at some point. He’s feeling left out, I’m sure. Yeah. Um, and we’ll just answer questions. Faster? Better? No. Yeah. [00:37:00] Are you familiar with a band called Sparks, Steve?

[00:37:03] Steve: No, 

[00:37:03] Steve: I’m not. 

[00:37:03] Cameron: God, man, you’re missing out. So, uh, Sparks is a band they’ve been around since.

[00:37:08] Steve: Oh, I’ve got to write that down. Every episode Cameron gives me something to write down, books to read and things to 

[00:37:13] Steve: check out. 

[00:37:14] Cameron: you, just missed it. 

[00:37:14] Steve: I spelled it with an X. 

[00:37:15] Steve: Should it be spelled with an 

[00:37:17] Cameron: A R K S. Two brothers, Ron and Russell Mael, M A E L. They’ve been, they’ve been around since 1971. They’re the most influential band you’ve never heard of. There was a great documentary that came out a few years ago by Edgar Wright about them called the Sparks Brothers.

[00:37:37] Cameron: You should check it out and watch it. Just look at the trailer, if nothing else. Anyway, they just toured Australia. Their first time back in Australia in like 25 years. Their first time ever to Brisbane and Chrissy and Fox and I went to see them. It was last week. It was an amazing concert. It was absolutely fabulous.

[00:37:57] Cameron: But all of their songs are kind of… [00:38:00] snarky social commentary, quite funny. And they’ve got a song. I can’t remember the name of it, but basically it’s, it talks about, I’m so confused. First, she said, your call is important to us. And then she said, please hold, please hold. It’s called, the song’s called mixed messages or something like that.

[00:38:20] Cameron: She said, your call is very important to us. And then she 

[00:38:23] Steve: I love that. I did a blog post on 

[00:38:24] Steve: that years ago. Your call is important to us. Except that it’s 

[00:38:28] Steve: not. Please hold. I mean, Yeah. 

[00:38:30] Cameron: we get away from that now. We have an, uh, uh, an always ready 24 7, uh, support bot that knows everything about the business and can answer all of your questions. But, I should go. Hey, Siri, let me talk to Maverick.

[00:38:43] Cameron: Maverick, tell me about the short term future for using generative AI tools in customer support, how that’s going to change customer support.

[00:38:55] Cameron: Generative AI is shaking things up in customer support. In the short term, we’re going to [00:39:00] see AI handling more of the basic stuff. I think resetting passwords, tracking orders, that sort of thing. This means

[00:39:07] Cameron: support reps can focus on the trickier cases that need a human touch. But it’s not all sunshine and rainbows. They’ll be growing pains. As AI gets smarter, businesses will have to balance using these tools without making customers feel like they’re talking to a brick wall. Uh, it’s all about the blend using AI to take care of the mundane while keeping that personal connection alive in customer service.

[00:39:30] Cameron: What’s your take on this? Well, thank you for asking. Um, yeah, I think it’s, I think it’s going to be dramatic and I think it’s going to be very fast. I think there’s going to be an expectation from customers that you should have this. I mean, it’s already my expectation and you know, bad customer support drives me nuts.

[00:39:51] Steve: it’s going to be like that. People will say, this is bullshit, man. I rang up for the AI and I got the person. I can’t believe this shit. 

[00:39:56] Cameron: I mean, humans, uh. Just [00:40:00] too slow. Don’t know. Stuff. Have to check. Oh, sorry. I have to go check that. Like really, I don’t want that. I, I mean, I, I want the ai, I, I need the AI to do my customer service. It can, it can escalate things to a human if need be, but AI will be a first, um, you know, port a call for most customer support, and I think it’s gonna do a much better 

[00:40:19] Steve: It’s one of the key things I talk about on stage is the dichotomy economy. Everything’s really moving to the edges at the moment. I mean, in retail, it’s Tiffany’s and Costco. It’s full service. It’s no service. It’s high price. It’s free. That’s kind of the companies that are really succeeding tend to have a position that lives on the edge.

[00:40:36] Steve: And I really like the idea of saying, AI is great because so, so many interactions that we have, we actually just want the thing solved and it’s nothing personal. And in fact, I wouldn’t want to waste a person’s time with something that is inconsequential. Like I was just fixing, um, before my arcade machine, the power source, um, blew up.

[00:40:56] Steve: So I have to get a new power source. And if that was something I could ring and I do it and there’s [00:41:00] an AI that says, this is the part you need. Here’s how it goes. You know, I don’t want to waste a person’s time on that. I want people to be redeployed to something where the humanity really matters, and then we can probably pay the humans more when they’re doing a human thing, because the AIs…

[00:41:15] Steve: Really low marginal cost to operate. And we get those things done that there’s a lot of pieces in the puzzle and processes, but now I can go through those real quick, far better than a human can. And then leave the humans to the tasks that really require empathy. And, and, and even just knowing that it’s a human, even if I always can have empathy, sometimes you just want to know that the person on the other end understands and feels, 

[00:41:39] Cameron: Yeah. Well, look, I think customer support’s gonna be one. early way that most organizations are going to have to deploy this. Um, And I think another is just in, uh, sort of analytics, business analytics. I mean, there’s just so [00:42:00] much room for faster and better analytics out there. Not just in terms of. You know, figuring out, you know, uh, where your revenue lines are and that sort of stuff, but really looking for new opportunities in the marketplace by grabbing huge amounts of data and having the AI sift through it, looking for the opportunities for your business, looking for.

[00:42:23] Steve: think so. 

[00:42:24] Cameron: ways to improve profit, reduce cost, improve customer service, etc.

[00:42:30] Steve: Data science was, it’s been a really trendy thing, really only for the last decade or so, a little bit more than that. But it almost didn’t exist before that. You wouldn’t have a data scientist, you might have a business analyst, but then data scientists came in. But I think it might be one of those roles that really only had a small window of existence.

[00:42:46] Steve: Because AI is the perfect thing to replace a data scientist. Because someone can semantically code it and say, I want you to look at these markets, these things, and look for any correlations. Or, you know, build up information that a data analyst might take two days to do, and [00:43:00] an AI can do it in two minutes. 

[00:43:01] Cameron: Well, moving on, Steve, the other story I wanted to talk about before we get into our deep dive is China is starting to play a bigger role in generating their own AIs. Alibaba this week launched the latest version of their artificial intelligence model, Tongyi Qianwen 2. 0. Probably pronounced that wrong.

[00:43:25] Cameron: You’ve got a Chinese speaking wife. Can you ask her how to pronounce that for me? 

[00:43:30] Steve: I’m a Chinese speaking person as well, but I’ve just got to, yeah. 我会说普通话, 很好, 因为我的太太是中国人 

[00:43:38] Cameron: Well, there you go. I did not know that. 

[00:43:40] Steve: know? I thought you knew. 

[00:43:41] Cameron: you? Are you, is it 

[00:43:43] Steve: Oh, I used to be pretty good. Now, 不错, 不错, that’s, that’s Mandarin?

[00:43:48] Steve: 不错, 不错, uh, which means not bad, not bad. I used to be pretty good when I used to travel there a lot for work. I’m just looking for this article here.

[00:43:58] Steve: Um, [00:44:00] Chinese AI, where is it? Tongyi Qianwen 

[00:44:05] Cameron: Hongyi Tongyi Qianwen 

[00:44:09] Steve: qianwen. Tungyi Qianwen. Qianwen. Wen means, um, language. The word wen means language and qian is, I don’t know what qian means. Tungyi, something, something language. My wife’s name is Shen Wen, which means sparkling language. 

[00:44:27] Cameron: Oh, lovely. Well, I’m just going to call it TQ. TQ. They also, Alibaba also introduced their Gen I service platform, which is apparently going to let companies build their own generative AI applications using their own data, just like GPT or OpenAI did today. So, uh, this is obviously part of the AI race is why I said earlier that I don’t necessarily think that open AI has got a motor, a lot of runway here.

[00:44:58] Cameron: There are going to be incredibly [00:45:00] powerful AIs coming out of Google, coming out of. Uh, Tess, uh, you know, X coming out of China, coming out of other countries I’m sure as well, but particularly I expect to see a lot China, a lot, a lot of this sort of technology coming out of China in the next couple of years.

[00:45:17] Steve: But it does seem though, that certain firms or race, and there tends to only really be one or two big winners in platform economies, you know, like with search, it’s turned out that way, with app stores, it’s turned out that way, you know, with hardware, with social, there tends to be a few winners and they tend to have a certain proclivity.

[00:45:36] Steve: To do well in a certain thing, you know, like Amazon is clearly a logistics company. That’s their, that’s their regime. That’s where they play. You know, Apple is all about, you know, UX. It’s always been their thing is, you know, deep down, Apple is all about UX. Google is a really scientific company. That’s why they’ve never really succeeded in social.

[00:45:53] Steve: They just kind of, their DNA doesn’t have that. Inside them. Do you know what I mean? So I do think that certain [00:46:00] companies have a proclivity or a DNA to develop things just because they have a certain attitude that permeates through the products that it puts in market. Um, and, and it would be interesting with LLMs and generative AI.

[00:46:13] Steve: Given that they’re all based on the same kind of idea, I wonder if they can extricate themselves from the company’s DNA that launches it. But the one thing that came to my mind with Chinese tech giants is, you know, what does their LLM train on? Does it train on the wider web or the Chinese web? Did it say that in the article? 

[00:46:29] Cameron: I didn’t, but I’m waiting for Tencent to become part of this, you know, I, I imagined that, WeChat, TikTok, these sorts of, um, Uh, uh, platforms are going to provide a huge, uh, information stream for training Chinese AIs on.

[00:46:50] Steve: Yeah, definitely. And I think that, you know, the CCP would be smart enough to know not to train it. On if they want a good LLM on, on, on [00:47:00] their, uh, walled garden, you know, the Chinese firewall, uh, for an AI, because, you know, quite frankly, it’s, it’s, uh, there’s limitations on the information that it could be trained on.

[00:47:12] Steve: But, um. That, that’ll be interesting to see what happens, whether or not they have two versions of it. They have their international and their local version, because certainly they do that with TikTok. You know, ByteDance has very, very different versions. It’s been well documented that the algorithms of what you show a teenager in China is very, very different to what you show a teenager in America.

[00:47:32] Steve: You know, in China, it’s all science experiments and achievement and all of this stuff. And you know, in America, it’s like, let’s dumb them down, just put the, put the scales on dumbing down our, our competitors across the Pacific. 

[00:47:44] Cameron: Hmm. Yeah. Well, I mean, certainly it’ll be interesting to see how that plays out and, you know, how the politics of it from the CCP need to be, um, engineered for a Chinese versus a non Chinese [00:48:00] audience. But certainly it’s going to be a, it’s going to be a 

[00:48:03] Steve: and if there’s, I was going to say, if there’s rogue behaviours on a GPT, because one of the things with the GPTs is, as was documented early on when ChatGPT became a global phenomenon was, How many ways you could hack it and that you just simply cannot think of all of the ways that you can ask it to give information or designed not to give.

[00:48:24] Steve: And they have to shut things down as they go and iterate. And I wonder how the CCP will manage that given the level of censorship they have on some, some sensitive issues they have in their country. You know, like, you know, the Tiananmen Square 

[00:48:37] Cameron: Hmm. you want to just move on to the double dive? The double D? 

[00:48:43] Steve: double D?

[00:48:44] Steve: Okay. So, well, I just wanted to get your thoughts on, in Australia yesterday, we had one of our, our second biggest telco went down for 12 hours or so. 10 million people were without internet access on their phone. It was, I got called up by the radio and TV to make some [00:49:00] commentary on it. Are we too reliant on technology?

[00:49:02] Steve: And I didn’t give them what they want. I said, look, first of all, let’s just, it’s an incredibly rare event. I don’t know if you remember, but electricity strikes, uh, blackouts used to happen all the time. And yeah, life went on and we didn’t say, you know what, this electricity is too dangerous. Let’s, let’s start lighting open fires again and getting candles because yo.

[00:49:21] Steve: So I didn’t say that but I said, look, it is important to have redundancy in areas where it matters. You know, like hospitals and emergency services and so on. And it would probably be valuable to have adversarial interoperability, uh, with things like SIM cards and so on, which we don’t have in this country, but you could have it in other ones. 

[00:49:40] Cameron: that? mean?

[00:49:41] Steve: And then the, 

[00:49:41] Cameron: That’s a new term to me.

[00:49:42] Steve: so adversarial interoperability. It’s kind of the idea that you have one platform that all technologies use. You know, the simplest, uh, example is the plugs in the wall. It tends to be the same in a country. You don’t want to have different companies with different plugs or the same software application that everyone can [00:50:00] understand.

[00:50:00] Steve: There’s been a lot of calls for you should have that in social media, so you can take your audience with you, that type of thing. And it’s kind of creates. The world that we live in, you know, HTTP protocol is adversarial. Anyone can use the same one. And so those walled gardens of having protected systems where one thing doesn’t work in another is what companies want because it affords them financial protection.

[00:50:24] Steve: And telcos are one of the classic ones. For a long time, you couldn’t transfer your phone number. And then we said, Hey, wait a minute, people should be able to keep their phone number if they change. So they changed that. But there’s another call for saying, well, we should have towers and SIM cards that can go across, you know, just with a flick of a switch to a different provider.

[00:50:41] Steve: And some countries have it. I think New Zealand and a couple of the Nordic countries have it where if there’s an outage, it can go across to your competitor. And it would be good if you could do that because you could do real time changing, you know, whoever’s offering the best price on a particular day, same with electricity, you should be able to do it, but you can’t.

[00:50:56] Steve: So. A lot of people were ringing in. I went on [00:51:00] three AW um, which is a callback station. It’s a number one rated station, but it has a lot of, uh, Luddites sort of saying that’s why cash is king. You gotta come back at cash ’cause your people got caught out. They couldn’t use their Apple pay. You couldn’t use your Google maps.

[00:51:15] Steve: And, um, I mean, the question that I just wanted to ask you can get your ports, your, your thoughts on it, you know, is a full digital world safe? I mean, I know that with military institutions we have air gapping and we have safety, you know, requisites, but not so much in consumer technology. 

[00:51:32] Cameron: Yeah. Well, the first thing I thought about was the electricity analogy as well. Look, the, the world, the modern world is built on technology. I mean, it is going to be, and there’s nothing… We can do about that. The people who say that we are too reliant on technology are obviously just having a go.

[00:51:53] Cameron: Um, it’s kind of a stupid statement from the get go, but we do need to get smarter at [00:52:00] building safety mechanisms and redundancies into these things. I think there should be penalties for corporations that… Uh, don’t, like, it’s the same with when they get hacked. When they get hacked, information gets stolen, there should be massive consequences for that.

[00:52:19] Cameron: Not just, uh, stern talking to, like they get today. There should be massive consequences. That’s, that’s… It’s a 

[00:52:27] Steve: are right. I. 

[00:52:27] Cameron: corporate responsibility to take all of this customer data and then allow it to get stolen because you didn’t work hard enough on building protection mechanisms into place.

[00:52:39] Steve: Yeah.

[00:52:40] Steve: I think that is exactly the number one thing. I mentioned that on the radio. Yes. I said, this should be compensation. If you’re in a business and you can show that you lost X amount of dollars because of the outage, then the company should pay. It’s like straight up, like a straight up, and you’ll find that these things happen far more rarely if one of those outages cost Optus, you know, 1.

[00:52:58] Steve: 3 billion in lost sales, uh, [00:53:00] because however many small businesses couldn’t take, take money from customers. I mean, and that would be pretty easy to prove. Wouldn’t it? Cause you just have to show the history of payments on the same day and the same year, and then just go, and this is how many we missed out on.

[00:53:11] Steve: So, you know, pay up. And it’s the same reason with data hacks that no, there’s no consequences. And when there’s no consequences, there’ll be no investment. 

[00:53:19] Cameron: Yeah, there needs to be consequences for organizations that fail to deliver the sort of uptime that is expected. You know, if we, if, if you’re selling me a service that, and you’re selling me on the fact that it’s, uh, that I can rely on it, then I need to be able to rely on it. Anyway, I, look, I think it was, it’ll be interesting to see how this plays out with Optus and the management there, including, I think, the former Premier of New South Wales isn’t, uh, the disgraced Premier of New South Wales, isn’t she one of the Optus senior management now?

[00:53:54] Cameron: That’s… 

[00:53:56] Steve: Yeah, yes, yes. You, you’ve seen your manager, [00:54:00] Gladys. 

[00:54:00] Cameron: it’ll be interesting to see how that plays out. I’m sure Gladys will be in overtime, uh, making phone calls in her little black book to shore up political support for the organization, but you know, getting back, this gets back to the story that we were talking about earlier when we are using artificial intelligence to run our businesses and run our lives and do all of our thinking for us.

[00:54:25] Cameron: And it goes down. And apparently OpenAI’s problems in the last 24 hours have been caused by a DDoS attack. They, they’ve had a, yeah, I saw, I was looking at their update page today, it was a DDoS attack on OpenAI. Now you have to wonder, who was doing that? The day after their big dev day and their big release of all of their new tools, they get a DDoS attack.

[00:54:53] Cameron: You’re 

[00:54:53] Steve: your bet Cam? You’re, you’re, you’re more experienced than almost anyone who [00:55:00] understands corporate chicanery and interoperations between countries and, you know, cold war. like,

[00:55:07] Steve: you’re the man for this one. 

[00:55:08] Cameron: Well, look, there are, there are two obvious candidates. Uh, it’s either somebody who just wants to watch the world burn, like an anonymous type organization that just decides AI is a dangerous thing and we’re just going to start to fuck with the infrastructure and make it difficult for people, cost them time and money and, and reduce the sense of reliability.

[00:55:33] Cameron: Or it’s a, an international player. Um, Unfortunately, um, what’s his, what’s his face? The guy who ran Russia’s private military. Who got blown up in his own jet. He was, he was, 

[00:55:49] Steve: Oh, that’s right. Yeah. I forget his name. 

[00:55:51] Cameron: um, he was running, uh, the, the operation, I think out of St. Petersburg that was supposedly doing a lot of this stuff for years. [00:56:00] Um, you know, it’s either China or Russia or, you know, North Korea, et cetera, et cetera, always the names that get thrown around, but really, uh, can they be bothered doing this kind of stuff?

[00:56:11] Cameron: I don’t know. The third option is one of their competitors. You know, it, maybe it’s a, it’s a Grok, or it’s a, it’s a, uh, Google, or somebody. 

[00:56:21] Steve: think any of them would stood to that. Oh, look, I can’t help but think that. You know, coincidences are so interesting because they’re coincidences. 

[00:56:29] Cameron: a DDoS, a DDoS attack’s not a coincidence. If somebody’s targeting you, 

[00:56:35] Steve: Oh, yeah, I guess you’re 

[00:56:36] Cameron: system, um, 

[00:56:38] Steve: yeah, you’re right. 

[00:56:40] Cameron: So, somebody deliberately attacked them yesterday, the day after their big announcement, when their systems would have been getting maxed out anyway, somebody hit them to try and do 

[00:56:52] Steve: Could it just be, could it just be the max out because, because it, could it just be a max out because the launch was,

[00:56:59] Steve: [00:57:00] so interesting and did so well, I’m just looking, Sam Altman’s keynote didn’t really go viral, viral. Would it get 800, 000 views? It’s not huge. I mean, it’s big, but it’s not, you know, blow the world up viral. 

[00:57:11] Cameron: Well, no, look, my initial assumption, I think their initial assumption too, probably would have been that it was just the backend systems were getting hammered because everyone was testing out the new tools. There’s always going to be a, a peak in people playing with this stuff in the first few days after an announcement.

[00:57:29] Cameron: We saw that when GPT first came out, there was a massive peak of interest for a few months and then it waned off and the media was like, oh. The GPT buzzes over like it’s, it’s all died off. But then they came out and said it was a DDoS attack today. So, um, yeah, somebody’s attacking him. I, I got no idea who it’d be.

[00:57:49] Cameron: My guess,

[00:57:49] Steve: don’t think it would. I don’t think it’s corporate espionage, corporate, I just don’t, I don’t know, maybe I’m wearing rose colored glasses, but I just don’t think that a public company [00:58:00] or a real competitor would do that necessarily. I just imagine that that would happen, but maybe it would, 

[00:58:07] Cameron: oh, I, I imagine corporate espionage happens all the time, Steve. But, uh, they would probably, if you’re going to do something like that, you would do it obviously through a series of fronts. So it would never come back. There’d be plausible deniability. But my guess is if I had to 

[00:58:24] Steve: but it would have to be pre planned. I’d have to be pre planned a DDoS attack, wouldn’t it? you couldn’t do it that quickly,

[00:58:32] Cameron: oh Yeah, you probably 

[00:58:34] Steve: How quick could, you deploy one? 

[00:58:35] Cameron: it’s just a, it’s just a bunch of bots, right, that you send out to hammer a destination. But my guess is that it’s probably an activist organization that are anti AI, that, uh, and I imagine…

[00:58:51] Steve: And I love that they use a form of AI to be anti AI. That’s what I love the most. That, that I love is the circular 

[00:58:58] Cameron: you watch Mr. Robot?[00:59:00] 

[00:59:01] Steve: I did. 

[00:59:03] Cameron: It was a great show until it sort of jumped the shark, I thought. 

[00:59:06] Steve: It lost its way a bit. Yeah. It lost its way a bit. They all do. It’s very hard to 

[00:59:10] Cameron: The first couple of seasons were great, yeah, yeah. But, uh, that’s who I imagine is. I think it’s Mr. Robot, somebody like that, trying to, you know, stop the system in its tracks. I don’t think anyone’s taken credit for it, though, which is unusual.

[00:59:26] Cameron: You’d expect that if it was an activist organization, they would have come out wearing their Guy Fawkes masks and, you know, taking credit for it. But we’ll 

[00:59:35] Steve: Yeah. 

[00:59:35] Cameron: it plays out.

[00:59:36] Steve: Yep. 

[00:59:37] Cameron: Technology Time

[00:59:38] Steve: Okay. Well, I just think that. I did a, uh, computer science course in 1992 at university, and one of the key things that they were talking about that would arrive then was the EMDS, the Executive Management Decision System software. And we would all be using by the [01:00:00] time we graduated, Cameron. We wouldn’t really be asking any questions, we’d be pumping these into a system, which would look at all of the data, give us little, they didn’t use the word the applications, but something like that, where it would synthesize all of the data, deliver it to us, and give us things that we need to make decisions upon.

[01:00:16] Steve: And I just thought it was such an interesting Time Warp, given, and if you read any of the old, uh, software textbooks from way back then, they all had that in there. And here we are, 30 years later. And, and it’s arrived, you know, like here they are with the GPTs coming online, which are going to be like little micro executive management decision system systems and tools to get things done.

[01:00:41] Steve: But that was the big promise. And this was kind of pre public web and pre browser. So at the time it was all about. Bespoke software going into your internal system, extracting data from your internal system, things from the business, the operations, the logistics, the manufacturing, the retailers, and pulling it all together.

[01:00:59] Steve: And [01:01:00] I just thought, wow, that’s so interesting that

[01:01:02] Steve: so many ideas are long in the tooth and And, and they take, they germinate for a really long time before the technology catches up to the promise of what we can imagine. And it just felt like a moment where that 

[01:01:14] Steve: time warp finally arrived.

[01:01:16] Steve: And I was watching the OpenAI presentation going, wow, this is kind of the promise of internal systems because it was pre web. It was no wider web. It’s like, you need a decision system on internal stuff to make things and do things internally. And so we finally got there 30 years later. 

[01:01:34] Cameron: Yeah.

[01:01:34] Cameron: Like, so one of the interesting things about the Dev Day announcement was the. Increase in the size of the context window now that ChatGPT has and that the API has. I think it’s gone from 8, 000 tokens to something like 128, 000 tokens or something. I remember Sam said, you can now. Uses like 300 pages of a book, [01:02:00] 

[01:02:00] Steve: three hundred pages in the book. 

[01:02:01] Cameron: which is pretty impressive, but looking at the AI Assistant stuff, where you can upload 20 documents, and I think each document has to be, uh, under 500 meg, I think.

[01:02:15] Cameron: Let me just check that. It might be 500 gig, hold on, which is a lot of documents, uh, 512 meg each, there you go. The size of all the files uploaded can’t exceed 100 gig, but it’s still a lot of files, right, if it’s 

[01:02:31] Steve: A lot. If you have the right ones, I mean, the problem that you might have, especially in corporate is that there’s so much repetition and the same things and having quality databases is becoming really important. Everyone’s databases are terribly unclean. 

[01:02:46] Cameron: But I was talking to Chrissy about this the other night, about whether or not books will be a thing anymore. And I’m talking about non fiction books as opposed to fiction, like you and I have written a fair share of non fiction books, and books [01:03:00] are great, I love books, I read books everyday, but it’s a linear experience, right, as you’re reading through it.

[01:03:09] Cameron: Um, now I’m not sure I want to read a nonfiction book anymore. Um, now if you took a book, one of your books, and you made a GPT about it, what was, what was the title of your last book?

[01:03:22] Steve: The lessons school forgot. 

[01:03:24] Cameron: So if you gave me that as a GPT. And I could ask you questions about it. Okay. So what’s this book about? Oh, okay. Tell me the top five, you know, uh, lessons that I should learn from this book. Oh, that’s interesting. Let’s drill down on that. Tell me a little bit more about that and why that’s important. And it becomes my own little, I don’t know, Wikipedia version of your book that talks to me using GPT voice.

[01:03:53] Cameron: My last book, The Psychopath Epidemic, you know, you can have a conversation with the book. [01:04:00] And you can say, okay, read me a 

[01:04:01] Steve: Conversation with the book is actually, that’s a really, really cool idea, Cam. Just pause on that, a conversation with a book. So could you have a conversation with a biography? Because then you might be talking to Arnold Schwarzenegger or Steve Martin or whoever you’ve read about. 

[01:04:19] Cameron: Yeah. I, 

[01:04:20] Steve: That becomes, that becomes an interesting 

[01:04:22] Cameron: well, a conversation with them or a conversation about them with the book. Um, you know, the ability to, as I told you at the beginning of this, like I uploaded the QAV book that I’ve been writing and I can interrogate the system now about the book and get it to flesh things out and it, and it.

[01:04:44] Cameron: Explain things that weren’t in the book too. Like when I was asking it about price to operating cashflow, it went off and started giving me definitions of price to operating cashflow and how operating cashflow is different from free cashflow and how price to operating [01:05:00] cashflow is different as a metric to price to earnings.

[01:05:02] Cameron: And, you know, it was pulling stuff that I know. Uh, in my notes, uh, it was getting it from outside. So you have the opportunity now to start to have conversations about a book. I, I could see a future, Steve, where you and I don’t publish a book, we publish a GPT. I’ve just published my new GPT 

[01:05:23] Steve: I like that. Yeah. 

[01:05:24] Cameron: on Subject X on Psychopaths, and here’s the link to go to the website, or here’s the app.

[01:05:32] Cameron: And you can interact with the GPT and learn all about this subject that I’ve just written about for you, you know?

[01:05:43] Cameron: Uh, so Cameron in the editing booth here, folks, literally, as soon as we stopped recording today, I created a psychopath GPT. I jumped on GPT and saw that they had actually launched the, my [01:06:00] GPTs functionality today. After they’ve gotten over their D dos attack. So I uploaded the PDF of my book. The psychopath epidemic turned it into a GPT within a minute. And then, uh, published it on Facebook, say, Hey, you want to know anything about psychopaths or what I think about how psychopaths run and ruin the world. Ask my GPT. Um, so if you want to check that out and have a play with it, you need to have a chat GPT plus membership. But go to the link. Uh, for this blog post for this episode, go to the blog post for this episode on futuristic pod.com. And have a play with it. 

[01:06:41] Cameron: Let me know what you think.

[01:06:44] Steve: I mean, it’s a really, I’ve really liked that a lot because I’ve been thinking about writing a book. Uh, I’ve got a couple of ideas, but I’m just like, is anyone going to read a book? And is it worth promoting the book that hard? You know, because I think things have really, really changed since I did my last one, which was [01:07:00] five years ago.

[01:07:01] Steve: Like it’s changed a lot in that time and 

[01:07:05] Cameron: Mm

[01:07:06] Steve: to publish a GPT, I mean, I’ve just written that down. I’m like, this is a seriously good idea. On an I, especially on something that would be a nonfiction book, it feels really strong because the book becomes, again, nonlinear. You ask it, you interrogate it, you, you put pieces in there.

[01:07:23] Steve: I was just thinking too, with a book, you gave me a great idea. I’m like, you could create a GPT that takes any book and you just simply say, give me different social media scripts. From that book. It doesn’t have to be the exact words, it could be whatever. Give me Instas script, give me a TikTok script.

[01:07:39] Steve: Give me, you know, with the visuals and everything goes through the book. And, and it does that immediately. I could do that with my lesson school, forgot, which is one of the things that I was thinking of doing. And now I could go in there and go, all right, here’s my book. Here’s, uh, the PDF of it all. It’s lesson 300 pages and, and create AGPT, uh, that turns it into books.

[01:07:58] Steve: So you could do it for anyone. You could book talk. [01:08:00] Here we go. 

[01:08:01] Cameron: Well, I’m thinking about something that’s part book, part podcast, part GPT. So I’m going to take All of the transcripts of all of, let’s say, my podcast series on Julius Caesar. Take it all, put it, and publish it as a GPT. And then people who know nothing about it can start listening to it. All right, uh, just read me the first chapter or the, the, you know, read me the first episode.

[01:08:34] Cameron: And it’ll, you know, it can play, maybe play the original podcast. Um, it’ll play sort of half an hour and then you can imagine you’re in your car, right? You know, you’re listening to an audio book and you can say, hold on, stop. Um, explain to me more about why. This character, who was Sulla? Tell me more about Sulla.

[01:08:53] Cameron: I don’t understand Sulla. What’s, what’s Sulla’s relationship to Julius Caesar? What was Sulla’s, [01:09:00] uh, importance in the history of Rome? Uh, what was Sulla’s relationship with Marius? And it can just, based on the notes from my podcast, the transcripts from my podcast, it can answer those questions the way that I would answer those questions.

[01:09:16] Cameron: Or it can go out to Wikipedia or go out to the wider web and answer those questions. So, and then you go, okay, that’s enough. Go back to re, go back to playing the episode or go back to reading the book or whatever it is.

[01:09:27] Steve: Yeah, you can, you can bounce off. It’s like a web experience with an AI that taps into what you need as you, as you do it with verbal instruction. 

[01:09:34] Cameron: and at the moment you can do that, but you switch between, say, an audio book and you see, then you jump over to GPT, 

[01:09:41] Steve: forums and yep. 

[01:09:43] Cameron: I can see where this is going to be a new kind of platform where I’ll be publishing my podcast slash book slash GPT. That’s more of a, an interactive entertainment slash educational experience.

[01:09:59] Steve: [01:10:00] And this really taps into the where do tomorrow’s jobs come from? 

[01:10:02] Cameron: yeah, I mean, I’m not sure everyone’s

[01:10:06] Steve: does. It really, for me, it does. No, not everyone’s going to have it, but so no, but this is, this is one example. Of something that you or I could do. So then there’s going to be a number of other examples for trades people and whatever, where they’re just leveling up on top of the technology and this panic that jobs are going to go away.

[01:10:26] Steve: I’ve always thought it just creates more jobs than it removes. But the one you had then, I’m like, well, I should put all my stuff in there and, and develop a. You know, Wikimee, you know, maybe you could have Wikimind GPTs where if someone who’s got a lot of published content, like you Cam or myself, it’s like, just becomes an interactive tool on the way I think about certain things or any thought leader thinks and knows about all, uh, all manner of topics that they’re interested in you could really tap into that.

[01:10:52] Steve: Using this kind of method. Imagine if you build a GPT that anyone could point their webpage and their social media to. [01:11:00] And actually it creates a conversational bot for that person or an information extraction machine on what Cam Reilly thinks about various issues or Steve Sammartino on tech or investing or whatever it is. 

[01:11:11] Cameron: Yeah, because 

[01:11:12] Steve: so much opportunity in this. So 

[01:11:13] Cameron: when you ask GPT to answer a question right now, it’s giving you an answer that’s been, it’s based on its training and its training is, you know, this massive online database and a book database and things like that. Um, but let’s say I’m only interested in how Noam Chomsky would answer that question.

[01:11:32] Cameron: You know, I want to tap into Noam Chomsky’s, all of his published stuff and get his view on it. Noam, it’s like, he’s like 96 or something now. Noam Chomsky’s not going to be around forever. But his body of work will survive him and that will end up as a GPT and I’ll be able to, you know, when the next Israel Gaza war breaks out five years from now, Noam’s not around, I’ll be able to say, what would, what would Noam, how would Noam talk [01:12:00] 

[01:12:00] Steve: What would, what would he 

[01:12:00] Steve: say about this? yeah, no, but I think if, but for non famous people or people who’ve even just got a body of work that could be private for now, you could plug it in and then, and then make it public or it’s a really strong 

[01:12:13] Cameron: or if you’re, if you’ve got 

[01:12:14] Steve: to GPT, Wiki, WikiMindGPT. 

[01:12:17] Cameron: deep knowledge in any topic, if you’re a, an engineer or you’re a, I don’t know, a Shop fitter or whatever your skill set, a violin teacher. Chrissy was talking about this the other night. She’s like, if, well, if I sort of captured everything that I know about teaching violin, that I tell my students, you know, based on the last 25 years of being a violin teacher, I could create a.

[01:12:42] Cameron: An AI tool to teach violin. Like, yeah, absolutely. You know, there’s a lot of visuals in that, but you could create those as well. I just wanted to say, I asked, uh, I asked Zoom’s AI companion to catch me up. And it says, um, The meeting so far has covered various topics [01:13:00] including the concept of the dichotomy economy and how businesses are moving towards the edges.

[01:13:05] Cameron: The discussion also touched on China’s role in generating their own artificial intelligence and the potential for corporate espionage. There was mention of a DDoS attack on a company and the need for consequences for organizations that fail to protect customer data. The conversation then shifted to the idea of having conversations with books using AI technology.

[01:13:24] Cameron: Overall, the meeting covered a range of subjects related to technology, business, and AI. Not bad, 

[01:13:31] Steve: It was pretty good summary, isn’t it? It’s actually, it’s not bad It’s pretty damn good. Like, if you weren’t in that meeting. And you got that, you’d be, you’d be right up to date. In fact, you might even be better up to date than the people who were in the 

[01:13:42] Cameron: Who tuned out and were looking at their phones.

[01:13:45] Steve: yeah, who were just doing whatever, exactly.

[01:13:47] Steve: Just thinking about what time they can leave the office, 

[01:13:50] Cameron: wrap. Let’s wrap this up with your futurist forecast for Episode 17, Steve.

[01:13:57] Steve: my future’s forecast is really simple and we’ve sort of [01:14:00] mentioned it. I feel like OpenAI could really build a platform for the ages. The idea that you’re going to have GPTs that, as you mentioned, you know, could have 10 billion developers in the world, just the cognitive surplus and creativity on top of AI and what could come out of that to me is extraordinary.

[01:14:20] Steve: Of course, there’s a lot of other LLMs out there, but. I think this small lead that they’ve got where they’re sort of opening it up to others to develop. Is, is a really, really significant lead in the market. Um, and I, I just think, I mean, we’ll just spend the last 10 minutes talking about the things that you could do with GPTs that you’ve made.

[01:14:41] Steve: And I think that open AI is really on the precipice of doing something extraordinary, whether or not it’s open AI or Microsoft or, you know, whether they separate out. Um, but I really think it’s on the precipice of something extraordinary because of the platform nature of how quick they’ve got to that before the others. 

[01:14:59] Cameron: [01:15:00] Yeah, and Sam Altman did say in his keynote that they’re going to be launching that store in the next couple of weeks, next few weeks. Uh, there’ll be revenue sharing agreements with the creators of the GPTs who want theirs to be public. They’re doing the full App Store model. And, you know, one of my concerns with…

[01:15:19] Cameron: OpenAI is that Microsoft will at some point decide to bring it all in house. Um, I’m not sure how that works with OpenAI’s charter that it was created under, where it was supposed to be open source and all of this kind of stuff, 

[01:15:37] Steve: that charter since it used to be a non profit, but it’s not anymore. Microsoft has a 49 percent share and they get 75 percent of OpenAI’s profits until they get their 10 billion investment back and they get to keep their 49 percent 

[01:15:51] Cameron: Yeah, but didn’t they, I think the non profits still exist, but it set up a commercial entity underneath the non profit. 

[01:15:59] Steve: [01:16:00] Yeah. I don’t know what sort of corporate chicanery is going on there, but, um, I look, my understanding is they’re certainly not non for profit anymore. I don’t know how they’ve. Managed it, but I feel like it’s a very commercial entity now. 

[01:16:12] Cameron: Yeah, well I, I think they set one, I think the story is they set one up underneath the non profit, but the non profit still is the, the parent of the commercial, um, and I’m not sure what restrictions they have under that, but, you know, I think sort of the death of open AI would be if Microsoft took it all in house because they don’t have a good track record of, um, taking over businesses like that and then continuing to run them successfully.

[01:16:40] Cameron: I mean, they, they… They took over Xbox, they took over Halo, they took over Hotmail back in the day, and they’ve bought a number of businesses over the years, which have survived, but don’t, you know, Xbox has definitely, um, lost its leadership in the marketplace to [01:17:00] PlayStation now. Uh, Hotmail, obviously, who uses Hotmail?

[01:17:03] Cameron: I still have a Hotmail account, but I haven’t looked at it 

[01:17:05] Steve: I love it. When I get an email from Hotmail, I love it. I just love it. It’s 

[01:17:09] Cameron: the only people who send me hotmail emails are Microsoft, because I think it’s attached to my Xbox account and my Microsoft account. Um, yeah, so, I mean, obviously Satya Nadella isn’t Steve Ballmer. Um, he’s done a pretty good job there.

[01:17:24] Cameron: You have to commend him on what he’s done, particularly with the OpenAI investment. But, um, yeah, I fear the day when they decide to bring it all in house. And, 

[01:17:35] Steve: they don’t because it seems as though the way they’re developing it is, is, is doing really well. Surely you wouldn’t change it anytime soon. 

[01:17:42] Cameron: Yeah, man. But how many times have we seen that happen to startups over the last 30 years? Right. Uh, great 

[01:17:49] Steve: We’re not going to integrate it. We promise we’re going to leave it and we’re going to whatever until we change our mind 

[01:17:54] Cameron: yeah, yeah, yeah. We love what you’re doing. We don’t want to change a thing. We 

[01:17:57] Steve: I love with you. We’re not going to, wreck anything until we 

[01:17:59] Cameron: [01:18:00] yeah, exactly. I mean, yeah, corporate. Power structures and the psychopaths within them, um, tend to ruin these sorts of things.

[01:18:08] Cameron: But, um, anyway, I, it’ll be interesting to see how that plays out. It’s just been an incredible couple of weeks, Steve. Things just seem to be moving faster and faster. Every time we do an episode, it just seems like we’re on the, we’re on the knee of the exponential curve of the hockey stick. Don’t you, don’t you feel that way?

[01:18:29] Steve: Oh, more so than ever. I just, it’s very difficult to keep up and, and that’s why we need more AI. So we don’t have to. 

[01:18:39] Cameron: Well, I tell you what I want to, the challenge I want to set both of us for our next episode is to come up with a list of, uh, realistic, Short term scenarios where AI is going to play a significant role in businesses and government [01:19:00] departments, institutions, how it’s going to require dramatic changes in the way that they think and operate.

[01:19:09] Cameron: Because I really want to, I mean, that was part of my impetus for doing the show was to, for the you and I to sort of. Have the space to think through that sort of stuff and pick it apart and start to figure out where the opportunities and the threats are for businesses. So I think we should come up with a couple of those each for next week.

[01:19:31] Cameron: What do you think?

[01:19:31] Steve: Let’s do that. 

[01:19:32] Cameron: Thanks, mate. You have a good week. 

[01:19:34] Steve: You too. Thanks, Cam.