This week on Futuristic – Brisbane portrait prize allows AI, Neuralink human trials have started, Berkley AI let their robot walk around SF, Arc Search is Google’s worst nightmare, why electricity was the AI of the 1800s, and Cam’s 9 scenarios for how the Future of AI might play out (and Steve’s adds #10).
FULL TRANSCRIPT
Fut 21
[00:00:00] Steve: 3, 2, 1, go.
[00:00:12] Cameron: Welcome to the Futuristic Podcast, episode twenty-one. Steve Sammartino, as seen on the project, uh, this week talking about AI and the future. How you doing Sammo?
[00:00:26] Steve: I’m good. I was on there. I’ll tell you what you uh, do 45 minutes for a wonderful three sound bites of one sentence.
[00:00:34] Steve: Yes.
[00:00:35] Cameron: Yes. Well, you can get it All out on this show, Steve You can talk for as long as you like. What’s one thing of note you did? Uh, futuristically, uh, this week, Steve?
[00:00:46] Steve: I’m gonna break the rule and
[00:00:47] Steve: give you two. Uh, but one thing that, that I noticed was I’ve been following a lot of TikTok AI update people, which they’re my feed is flooded with. And every time I see one of those updates I favorite, I go, geez, that’s a useful tool. Geez,
[00:01:03] Steve: that’s a useful tool. And then I sat
[00:01:06] Steve: down to use a few of them. I thought, I’m gonna have a go at these. And I found that eight outta 10 of them are pretty crappy. Now, when I say crappy, they’re kind of okay, but they are far more manual than you think. It’s not just you type in, you do this and it does that, or you put up a video and it gives you 20 different TikTok sound bites and picks the perfect ones.
[00:01:26] Steve: The fact is it’s a Pareto principle. You know, 20% of them are good and 80% of ’em are really crappy and. It was just a, a, a bit of a moment where that hype cycle, everyone’s creating apps and promoting them, and when they have the promotional video, they look great. I still find that a lot of them, and I’ll use the word yet, yet, are not very good.
[00:01:46] Steve: I’m sure they’ll get better very soon, but my insight is that I can’t help but think that all of those are just gonna get munched away by Gemini and OpenAI who are just almost, again, it’s a little bit like what happened with, uh, APIs on Twitter and Facebook. A little bit like the app Store. Let the crowd do the work of the type of things that look interesting and good, and then just do it yourself and shut ’em down.
[00:02:10] Steve: Here we are again. History repeats. Cam.
[00:02:13] Cameron: Sam Altman himself said, uh, a couple weeks ago, I
[00:02:16] Cameron: think we talked about this on an earlier episode. He said it, I think it was at the Y Combinator launch. Look, uh, don’t build a business around, you know, uh, putting bells and whistles onto GPT. ’cause GPT five is just gonna make all of it, uh, unnecessary. So, uh, you know, he’s, he’s being upfront with people in that regard saying, you know, don’t do that.
[00:02:37] Cameron: It’s, you’re gonna be disappointed.
[00:02:40] Steve: but every app that I see, I’m like, yeah, that’s just gonna be in Open AI and gonna be better, or it’s gonna be in Gemini because most of them are just those little enhancements on video and voice and really
[00:02:50] Steve: just tweaking the edges, bells and whistles is a good way to put it. Um, and one other thing, I did a keynote yesterday for a solar company.
[00:02:58] Steve: One thing that I’m noticing, I do a couple of demos of how you can use GPTs, and the demos that I do are usually.
[00:03:05] Steve: How you put stuff in to get something out. This is one little thing that I don’t think people are doing much, is people have this habit of the internet where the internet has been for the longest time, a a giant filing cabinet.
[00:03:17] Steve: You pull things out of it, right? You’ve gotta hope that someone has made it and you ask it and it delivers that thing. And the shift is that the internet now is a giant brain. But what people forget with, I think with open ai, and this is wider society, they’re going to the AIs to ask it to do things and very, very few people actually putting something in so that it knows more to pull out the thing that you want.
[00:03:42] Steve: And I just think that that putting an input in there to make it work harder and more specifically on your particular problem is something that most people aren’t doing. So that was of note.
[00:03:50] Cameron: What do you mean by putting something in? Like in what way?
[00:03:53] Steve: So up, uploading your data, so your own PDF or your, other than video and image, but I mean just actually adding to the database, you know, in the immediate thing that you’re doing. I know that you and I do that, but it seems that most people are like, oh wow, so you can put your PDF in and get it to analyze this and your financial data.
[00:04:12] Steve: And it just seems, and these are business audiences and I just thought they’d be further along and they just seem to be going to
[00:04:19] Steve: it and asking and requesting, but not so much saying, oh, how can I add to this database with what I’m working on or what’s in my, you know, corporate enclave of knowledge or my personal cloud.
[00:04:30] Steve: It doesn’t seem like many people are doing that at this point. I just thought that was interesting.
[00:04:35] Cameron: Well, of course, you know, we are not able to upload stuff that goes into the training database, but you can. Give it your own documents. And depending on what you’re using, if, if you are, if you are running your own model, if you’re using GPT for all or something like that, and you’ve built and training your own LLM locally, which I’ve spent a bit of time doing, you can give it a ton of documents.
[00:04:57] Cameron: But even that process isn’t that easy. You have to chunk the documents down, you need to strip all of the unnecessary metadata out of them. It’s, it’s quite a big exercise. Um, but yeah, I think people have still got their training wheels on with this stuff, and that will be true probably for quite some time.
[00:05:16] Cameron: Well, this week, uh, well, since we last did a show, which was a couple of weeks ago, uh, I’ve spent a lot of time writing more iOS shortcuts using GPT’s help. Um, I have one, you know, weigh myself every morning, as you know, I. Wash my diet and my weight and my exercise and all that pretty carefully. I have to, because if I don’t, I just, you know, eat shit and put on weight.
[00:05:38] Cameron: And, um, I built one where I can just, I have an iOS shortcut on my phone. I can just open it up, type in what my weight is when I get off the scales in the morning, and it will update it both into the health app in iOS, on my iPhone, but also do a Google sheet where I track all of this stuff and it’ll automatically update the Google sheet.
[00:05:57] Cameron: So I have the data stored in separate places and I, I’ve, I’ve got one that logs my kung fu workouts. When I get out of kung fu, I can pull up a shortcut, tell it which kind of workout I did. Some are more, uh, high intensity than others. Some go for an hour, some go for 90 minutes, some go for several hours.
[00:06:16] Cameron: And it will again, update my calories and, uh, put that into both the health app and spreadsheets and track it. So just learning how to. You know, automate more things, track more data using Code GPT to help me build this stuff.
[00:06:32] Cameron: It’s kind of nerdy stuff, but it, it’s sort of an example for me of how we’re all gonna be living a couple of years from now where you’ll be able to use these tools to automate a bunch of things for you.
[00:06:43] Cameron: Track stuff that’s important for you personally. Personalized software or as secret software. As I said, someone I read a, I read A, I think it was a Corey Doctorow post a couple of weeks ago. He was talking about when the whole idea of Life hacks was launched in the early two thousands at a conference.
[00:07:02] Cameron: The guy who did the talk on it referred to Secret Software, which was software that. Coders write, developers write for themselves. It never sees the light of day. It’s just software that runs on their devices that they’ve written to Automate functions for themselves. They call it secret software. So I’ve started writing a lot of secret software over the last, uh, couple of months, which is fun.
[00:07:25] Cameron: Well, getting into the top three, uh, news stories, Steve, the big one that dropped this morning and seems to have blown everybody away, including me. It’s, it’s one of the, like, it it is the, I think probably the biggest jaw dropping moments since chat. GPT first dropped publicly, uh, a little bit over a year ago, and that is open AI’s, new Sora.
[00:07:57] Cameron: S-O-R-A. Text to video. We’ve seen a lot of text to video tools come out in the last six months. Usually they will produce a second or a couple of seconds of little bit, uh, nasty looking video, but from text. But that’s a, that’s a huge achievement. The preview that OpenAI dropped today of S-O-R-A will generate 60 seconds of jaw-droppingly.
[00:08:28] Cameron: Good video from a text prompt. And here’s the cool thing. I dunno if you had, how much you dropped down this on the website where they did the
[00:08:36] Cameron: preview, they’ve got a bunch of videos that they’re showing. It’s not open to the public yet. They’re red teaming it. So they’ve got a group of people that they’re trying to use to find bugs and problems and think about ethical issues, et cetera, et cetera.
[00:08:52] Cameron: They’re giving it to a small group of researchers to play with. But what’s Sam Altman’s been doing today on Twitter is saying to people, Hey,
[00:09:00] Cameron: gimme a text prompt and I’ll make a video for you. And so people are just feeding him
[00:09:05] Steve: What a genius promotion that is. Genius
[00:09:09] Cameron: can’t give you direct access, but you just give me your prompt and I’ll make the
[00:09:13] Cameron: video for you And so on his
[00:09:14] Steve: Now I’m,
[00:09:15] Cameron: Sam A, if people aren’t already on it, he’s just churning those.
[00:09:19] Cameron: I mean
[00:09:20] Cameron: he was just churning him out this morning giving people demos of based on their
[00:09:25] Cameron: text prompts. I was like, that is, that’s smart. That’s
[00:09:28] Cameron: cool.
[00:09:29] Steve: here we go. Here’s one That I now a bicycle race on. An ocean with different animals and athletes riding on bicycles with their own camera view.
[00:09:36] Cameron: a dolphin in a seal with
[00:09:38] Steve: today. It’s actually, it’s actually interesting.
[00:09:41] Steve: There you go.
[00:09:42] Cameron: what did you think when you saw the video that this is producing? Steve?
[00:09:47] Steve: It looked, looked really great. I mean the, the first. Some of them looked a little bit green-Screeny, like the one of the car driving, there’s one of a car driving down a mountainside. That one looked a little bit green-Screeny, the one that blew me away was historical footage of the Gold Rush. Kind of was as good as any movie set.
[00:10:05] Steve: It looked, looked,
[00:10:05] Steve: there, it was like a drone view as well. The girl working, walking in Tokyo. I mean, I thought it was extraordinary. I just,
[00:10:12] Steve: the first thing I thought when I saw it is how quick can I use this? Because there’s so much content that I produce that is verbal, right? And if I could just add a little video on top of that, it’s written in verbal, and then create visual with that, it’s, it’s pretty extraordinary.
[00:10:28] Steve: It really is. And and the other thing I thought straight away was the first thing that I mentioned, which is the big players are just gonna eat up everyone who’s fringe dwelling on a little bell and whistle. I mean, this is a classic example, right? There’s quite a few out there and then they’ve just gone, yeah, let’s 10 x what you guys have done with the three second little video you can generate.
[00:10:45] Steve: I.
[00:10:46] Cameron: I mean, we’ve talked about this on the show over the last year or so, but, um, you know, I, I’ve been, uh, predicting for a long time that the impact that this is gonna have Generative AI is gonna have on social media like TikTok and Instagram and those sorts of video platforms, but also cinema and TV industries, when you can
[00:11:11] Cameron: just do text to video of a
[00:11:13] Cameron: very high quality, I think it’s gonna have a huge impact on people’s ability to create programming at a very low cost. And, it was, um. Uh, What’s-his-face, uh, head of DreamWorks who came out recently and said it was going to reduce animation costs by like 90% in the next five years.
[00:11:35] Steve: definitely. Yeah. We talked about early on how you used to make a lot of bedtime stories for Fox using chat, GPT. And, and you know, this is the multimodal element coming in here. So all of a sudden you’ve got a young kid. Not only can you write them a story, you can get the animation of some visuals.
[00:11:50] Steve: It’s not hard
[00:11:51] Steve: to see where this will go, where you have a make me a five minute bedtime story, uh, with a unicorn, uh, talking horse that saves the world from climate
[00:11:59] Steve: change with, you know, Laura Sammartino. It’s like, there you have it.
[00:12:05] Cameron: Yeah, and like to think that a little bit over a year ago we were all, uh, picking our jaws up, up off the ground because we had a tool that was generating text. Here we are a little bit over a year later, they’re generating 60 seconds worth of jaw-droppingly good cinematic quality video from text is like, wow, where are we gonna be a year from now?
[00:12:29] Cameron: Now you mentioned bedtime stories. One of the other things that happened in the last week is. We’ve all been given access to Gemini Advanced, which is, uh, Google’s barred Generative AI tool is now called Gemini. Gemini Advanced became available. Uh, they, I think they’re giving a two-month free trial if you sign up to advance, but then it’s 30 bucks Aussie a month for the premium level.
[00:12:58] Cameron: I signed up to it and I had heard, I read this great, uh, Reddit post where somebody did a breakdown very quickly, like within 24 hours after it came out. Uh, testing, uh, uh, Gemini Advanced against GPT-Four Turbo. Uh, one of the things that they said they thought Gemini Advanced was better at was creative writing.
[00:13:24] Cameron: So I, uh, did go back into bedtime story mode, uh, gave both GPT-Four and Gemini advance the same. Very, very simple prompt. To write a bedtime
[00:13:36] Cameron: story. And then I
[00:13:38] Cameron: posted both of them to Facebook
[00:13:40] Cameron: and, asked people what they thought, uh, who they thought,
[00:13:44] Cameron: did the better job. And there was a range of
[00:13:46] Cameron: views. Uh, both of them still not great, but I thought
[00:13:52] Cameron: Gemini advanced probably, you know, it was less
[00:13:57] Cameron: cliche.
[00:13:58] Cameron: You know, GPT still, when it writes these stories, they’re very flowery, very cliched.
[00:14:05] Steve: They
[00:14:05] Cameron: Gemini’s was much more
[00:14:07] Steve: stuff is very and, and, and just like that as well. GPT is very MBA speak on its business stuff too. It actually lacks nuance in, in a, in a lot of those areas. So the stories are cliched, the business stuff, it gives you back is MBA wash, you know, like it’s,
[00:14:24] Cameron: can, but you can.
[00:14:26] Steve: that in all of.
[00:14:26] Cameron: You can train it with custom instructions, etc. Not to do that with the business speak, but I’ve spent a lot of time over the last year trying to train it how to write creatively and it’s uh, it’s sort of pretty much been a waste of time. Like it doesn’t matter what I tell it, it doesn’t get any better.
[00:14:43] Cameron: Gemini is better in this regard still. It’s not gonna win any literary awards, and I don’t wanna take away from the fact that. It’s fucking amazing that you can ask a computer to write you an original story and it then does it. That in and of itself is jaw-dropping in terms of the history of computing.
[00:15:02] Cameron: But, uh, yeah, still not, uh, blowing your mind. But, you know, uh, how long before they do blow our minds. I don’t think it’ll be long just running down this guy. This is Lord per maximum on the, uh, singularity subreddit from three days ago. I made a comparison post before based on the view of non Reddit people on two models.
[00:15:22] Cameron: After testing two models extensively in the last few days. I feel like I have to share my honest thoughts on this. First and foremost, GPT-Four Turbo is significantly better than GPT-Four, so I’ll only include that in comparison, says GPT-Four Turbo is better at reasoning and logical deductions. Gemini Advanced may succeed somewhere where.
[00:15:41] Cameron: Uh, at some where GPT-Four Turbo fails, but still, GPT-Four Turbo is better at the majority of them in reality, even Gemini Pro seems a bit better than advanced Ultra at this. I’m not saying a lot though, because if a reasoning test is not in their training data, all of the models are bad. They really can’t generalize says GPT-Four Turbo is better at coding.
[00:16:03] Cameron: GPT-Four Turbo Hallucinates less than Gemini. He gave Gemini advance a win for creative writing for general conversations. Says it’s about two to three times faster than GPT-Four Turbo has no message. Cap refuses to do tasks, though more than GPT-4. So he gave GPT-Four Turbo the win for that. Um. GPT-Four
[00:16:34] Cameron: does better with, uh, code interpreter and more, uh, languages.
[00:16:40] Cameron: So basically he still had GPT-Four, Turbo, uh, winning five
[00:16:46] Cameron: out of nine tests that he put it up against. Five out of you, five verse four. But you know, that’s, that’s a big thing for Google to be catching up, uh, on some
[00:16:57] Cameron: metrics there. So that’s one person’s tests, you know, take it for what it is.
[00:17:03] Steve: it brings me to, to, you know, the business side of this, which I think is gonna be incredibly interesting. Um, the one thing that I’m kind of happy, happy about is the internet might turn towards a subscription model instead of an advertising-based model. I mean, I really, I really do hope that for a whole lot of reasons, ’cause I think that’s the original sin of the internet, is the attention economy that’s spawned under it and surveillance capitalism.
[00:17:25] Steve: But what it does do is it, it makes me
[00:17:28] Steve: think about streaming and the business model with
[00:17:30] Steve: streaming. And as you know, subscriptions in many areas of life are just growing 10 x versus the S&P 500. I wonder who’s gonna be the Netflix, like the one that you just go, this is the one that I am, I am always subscribed to.
[00:17:48] Steve: And then who are the others? Like who wins the subscription battle? Because it feels like you might be subscribed to one and not the other, and you would get better at using one versus the other. And I, I really gravitated towards what the emergent business models are. Do you end up with these, this battle of AI subscriptions?
[00:18:09] Steve: How many do we end up having, you know, at home we’ve got Netflix and Disney, but at one point we had Amazon and, and Apple and, and Binge. And then you are going, ah, and you, you end up turning ’em off and just having one or two. So I, I’m really interested in the business model that emerges from this and who you end up subscribing to or whether they gravitate down towards that.
[00:18:33] Steve: Now let’s just make this free and make it available to everyone, you know, as the, the new models emerge. I dunno, what do you think?
[00:18:40] Cameron: I’ll get to what I think in a moment, but now we need to pause for an ad from Manscaped. I’m kidding. Um, we,
[00:18:49] Steve: I thought that was real. I’m like, who knew?
[00:18:52] Cameron: uh, yeah, look, it’s gonna
[00:18:55] Steve: made you say that? ‘
[00:18:57] Cameron: cause you were dissing advertising. I thought it’d be funny if I threw to an ad.
[00:19:01] Steve: No. Okay. They’re good. Yeah. Good, good,
[00:19:03] Cameron: Uh, yeah. Look, I, I, I agree with you. Advertising, uh, has turned the internet into a shit show. Um. I, I also wonder who’s gonna be the top dog. You know, I think that it’s funny when, when I wanted to test Gemini and I’ve been testing, you know, I test PI and I’ve test perplexity and I test Gemini and I go in and out.
[00:19:26] Cameron: I always come
[00:19:26] Cameron: back to chat GPT, partly because I think it’s the best experience, but also partly because
[00:19:34] Cameron: I’ve kind of built up a bit of a product loyalty. I’ve got, yeah, I’ve
[00:19:39] Cameron: got conversations, I’ve got exports. I think, ah, if I start
[00:19:42] Cameron: spreading my conversations
[00:19:44] Cameron: around, I won’t know where was
[00:19:45] Cameron: that conversation and uh, how do I find a backup?
[00:19:49] Cameron: How do I, you know, where do I go to get that,
[00:19:52] Cameron: uh, you know.
[00:19:53] Steve: not just loyalty, it’s a little bit like when social media was big, you know, 15 years ago I found that, um, people would say, which is the best one? And I used to say to them, ’cause I was big on blogging and Twitter, they were the two main ones for me. Um, you know, I got a lot of business value out of them.
[00:20:10] Steve: I was like, just choose one and get good at it because there’s nuances in being good at it. ’cause they’re all a little bit different and have different personalities and user experiences. And if you get good at it, you’ll get more out of it given that they’re all reasonably good. it feels like that again, to an extent.
[00:20:25] Cameron: I think you’re
[00:20:26] Steve: sort of get good at it. Yeah.
[00:20:29] Cameron: But, you know, I, I, you know, I, I think that I’m waiting for Apple to come out with one. I’ll probably, uh, have a hard time
[00:20:38] Cameron: not using an Apple one if it’s integrated into
[00:20:41] Cameron: all of my Apple devices. I also think that the one that has integration with my email, my calendar, my documents,
[00:20:53] Cameron: my messages, all my, you know, all my
[00:20:56] Cameron: life is probably gonna be the one that will get my loyalty, and that’s probably gonna be Apple or Google, I guess, with the ecosystem that I live in.
[00:21:07] Steve: Yeah, and, and that’s the thing too, is that. The ones that seem to be looking like, they’ll, they’ll do better in the, in the long run it’s gonna be one of big tech again, which is one of the arguments with, you know, the antitrust heroes that are out there fighting the battles against big tech. It’s like, well, even which one you choose, guess what?
[00:21:24] Steve: There’s a super high chance it’s gonna be one of the five big technology companies floating around.
[00:21:28] Cameron: Mm mm Well, moving on, one X Robots, Steve uh, posted a video
[00:21:37] Cameron: in the last week of their robots. I hadn’t heard of these guys before.
[00:21:43] Steve: Oh man. I’m either,
[00:21:44] Cameron: Uh, it’s pretty good video. I mean, it’s, uh, two and a nearly three minutes long. One shot video. Uh, these are humanoid robots with
[00:21:58] Cameron: I would,
[00:21:59] Steve: little wheels
[00:22:00] Cameron: yeah. They’re on wheels. They’re not walking.
[00:22:02] Cameron: They’re like on, um,
[00:22:03] Cameron: a uh, those, they’re on a segue. That’s what I was gonna say. Yeah. With faces that look like they’re out of an episode of Dr. who they’re just like, uh, LED eyes in a smiley mouth. Quite creepy. I, If I saw one of those in a Dr. who episode I’d expected to start killing people a minute later, they look so creepy.
[00:22:26] Steve: Or rolling towards you in a dark alleyway in Melbourne, after you’ve had a few brewski, you, you really wouldn’t walk up towards it and go to shake its hand, would you?
[00:22:35] Cameron: No, no, it’s very terrifying looking. Eric Jang from one X says, here’s our latest software update on one x ai. Every behavior you see in this video is controlled from pixels to actions with a single neural net architecture. No tele-op, no scripted replay or specific, uh, task-specific code. No CGI all in one continuous video shot.
[00:23:01] Cameron: Obviously, we’ve seen different situations where a lot of these videos have been faked or Elon. Musk famously had optimists. The Optimus laws person dressed up in a robot suit, come out and do a dance on stage. Um, Google’s uh, Gemini video was heavily edited Yeah, a couple of
[00:23:21] Cameron: weeks ago. Um, so this is an, I mean this is an interesting video.
[00:23:27] Cameron: Part of me is unimpressed
[00:23:29] Cameron: at how slowly these robots are moving and doing all of these jobs of sorting things and moving portal cubes into a box, etc. Etc. Although, I mean, I should be highly impressed that we have humanoid robots doing anything. What did you think of
[00:23:46] Cameron: this video when you saw it?
[00:23:49] Steve: For me, it really showed how quick this is gonna change. And I’ll just pick up on the point you say unimpressed, they’re moving so slowly, but you know, doing things like picking up boxes and stacking them, looks like they were doing envelopes in a, almost like a mail sorting kind of setting
[00:24:07] Steve: or a office warehouse sort of
[00:24:09] Steve: setting. I mean, I the first thing I thought of, if they can do this
[00:24:12] Steve: now and it’s no code, it’s pixel to straight doing it and it’s
[00:24:16] Steve: not gamed, I’m like, imagine where they’ll be in 12 and twenty-four months. The first thing I thought of, you know, the ability to, I dunno if you can teach it visually. Um, some of the bots that I’ve seen before, you can actually, you show it what to do physically and then it copies you and it knows how to do that.
[00:24:34] Steve: Um, I just think that the ability for these things to do physical labor in warehouse office, factory settings, um, is gonna be extraordinary in two to three years. You just think about that, there’s almost nowhere that you couldn’t use it. Um, so I, I was, I was impressed and I just gravitated to remember exponential, Steve, it doesn’t get a little bit better.
[00:24:58] Steve: It gets twice as good and then twice as good as that, and then twice as good as that. And it almost doubles back to where we were, you know, just getting super excited about text. That made sense. You know, 18 months ago on the first chat GPT and now we’re talking video and here’s a, a rolling soft
[00:25:13] Steve: robot. I mean, Kurzweil is right.
[00:25:16] Steve: It just feels like, imagine where we’ll be in 2029.
[00:25:21] Cameron: I agree. I think it’s, uh, you know, very impressive at that level. But I watched an interesting video. So there was this thing, I think it was in the UAE, the World Government’s Summit held this week. Uh, it was a lot of discussion about ai. Sam Altman appeared over a video link and gave a talk, which was interesting.
[00:25:42] Cameron: And again, he talked about, uh, how much smarter GPT-Five is going to be and how they’re teaching it, reasoning, and it’s, it’s gonna just blow everyone away. But then Yann LeCun, the chief AI scientist of Meta, was on stage and was part of a panel and an interview. He made some interesting points about robotics and, and generative AI and some of the hurdles that we’re facing.
[00:26:11] Cameron: He, he said at one point that if you have a, a human baby that is 10 months old, has an intuitive understanding about how gravity works. They understand that if an object is dropped from a height and it’s unsupported, it’ll fall to the ground. But we still don’t have robots that can understand that. Um, and he was talking about how you can only train a.
[00:26:37] Cameron: Robots and AI using text up to a certain level and that we’ve run out of text. So, you know, we’ve al already training these LLMs on the entire internet. There’s no, it’s not like we can get a a hundred more Internets to train them on. We’ve trained them on the entire internet, publicly available internet information is talking about how we need to train them on video.
[00:27:00] Cameron: He said like, I think a four-year-old child has seen 16,000 hours of video. That’s about 30 minutes of what gets 30 minutes of YouTube up. YouTube uploads equals about 16,000 hours. So we’ve got a ton of video that we can start training AIs on, but as he was saying, it doesn’t work. You know, the, the way that we build generative AI on text doesn’t work on video.
[00:27:28] Cameron: So we need some huge breakthroughs in how we train AI models so we can train them on multimodal stuff more effectively, particularly video, and therefore be able to train robots far more effectively. So he’s pouring a lot of cold water on how quickly this is going to move into, you know, a, a, a much more highly advanced.
[00:27:54] Cameron: So basically saying, look, the, you can’t just throw more hardware at this and expect it to be much better than it already is because, you know, there’s not more data to
[00:28:04] Cameron: train it on. More hardware isn’t gonna make it significantly better.
[00:28:10] Steve: Oh, two things came to mind when I read the article was
[00:28:14] Steve: in some ways if there is a physical limit, that might be good. It might be good that we get to a certain limit where there’s a certain capability and, and things top out because
[00:28:23] Steve: we have this view that things will consistently be get better. And I always come back
[00:28:28] Steve: to you know, aircraft, you know,
[00:28:30] Steve: in 1950 it was like, you’ll fly
[00:28:32] Steve: in an hour to London.
[00:28:33] Steve: And you know, even beyond Supersonic, there was all of this, it’ll just continue to get better forever. And it, and it didn’t. And when we just moved to making the price better, more efficient, and better human, you know, UX, let’s say in the plains and so on, it’s gone the other way in some areas. But that’s kind of where that industrial and flying technology went to.
[00:28:53] Steve: And that might be a good thing that this happens with robotics. But the one thing that I see all the time with AI experts, when they make comparisons to biological beings, you know, not as smart as a cat, doesn’t have as many neurons and doesn’t have therefore a view of the world. It, it’s almost like they forget.
[00:29:10] Steve: It doesn’t need. The, the number of neurons that we need to do extraordinary and even dangerous things. I mean, any animal needs to know how to gather food, how to breathe, how to do all of these things just to keep its body alive. What’s don’t need that. They only need a certain slither and they only need to be exponentially good in a certain area to become dangerous.
[00:29:31] Steve: And you know, of course there’s, you know, cats and dogs and birds have their own objectives and they’re all self-directed. We’ve discussed this before, and while bots may not possess that, if they get super smart in a certain slither. might be enough for things to be dangerous. So there are two things that I thought about and I, and I do think that sometimes technologists, and it’s the same with lawyers and engineers and marketers, you get too close to what you do to have a, a broad view or perspective of risk and how things might emanate.
[00:30:01] Steve: And we’ve seen that with big tech, you know, denying, well, they’re probably just denying it, but sort of saying, no, it’s not this, it’s not that. And we’re like, well, actually it is. And there’s always a danger with someone who works in the technology who just cannot see the forest for the trees with the way that they look at things because they are so deep inside a particular context that they themselves, ironically, lack the worldview.
[00:30:23] Cameron: Sure I take your point, but I think Yann LeCun probably has a good sense of what the limitations are in terms of training AI models. Um, and, and I, and I’ve been saying this for a long time myself on the show, that I do think people, uh, tend to think that the LLM approach to generative AI is the be-all and end-all solution
[00:30:49] Cameron: for the artificial intelligence
[00:30:52] Cameron: journey. I don’t, I think we do need, uh, different systems, expert
[00:30:57] Cameron: systems, systems that use symbolic logic, but I do think the LLMs are gonna help us get there, that they will be able to help us, uh, uh, magnify the research and development process that might get us to the next level.
[00:31:14] Steve: Yeah, the next curve jump usually invents the next technology and the next curve jump. I mean, the thing that I just keep coming back to is nowhere near human level intelligence, but that doesn’t need human level intelligence for it to be significant and or dangerous and or, you know,
[00:31:28] Cameron: Particularly when you just need to train one robot how to do something, and then you can download that model into another billion robots in a minute and then they can all do it. Right.
[00:31:38] Steve: Exactly. Exactly.
[00:31:39] Cameron: I know we’ve done our three stories, but one that I didn’t put in the notes, we have to mention briefly. The Vision Pro, Apple’s Vision Pro actually hit the market in the us.
[00:31:48] Cameron: One of my, well, both of my older boys are in LA at the moment, and, uh, one of them, Taylor went and had a 10 minute demo at an Apple store of the Vision Pro. Um, and yeah, he said it was amazing, like mind-blowingly good. All the usual stuff that we’ve been hearing. If you’re following people on Twitter or Reddit or YouTube, um, Marcus, uh, et cetera, who have been testing these things out very heavy on the head.
[00:32:20] Cameron: Scobar was talking about how his gives him a headache because of how much his eyes are working and he’s not blinking. You know, there’s a lot of, um, practical problems with using them. That said, everyone, including David Letterman, there’s a great video of David Letterman testing one out and just, just blowing his mind is how, is what a huge leap the, uh, experience is in terms of interacting with the technology and the possibilities of it.
[00:32:51] Cameron: The three, the, the. Panoramic views, the 3D views. Taylor was saying he watched some 3D movies and he said it’s just like, it’s mind
[00:33:00] Cameron: blowing the 3D experience on it. So it’s out. I, I agree with the perspective that
[00:33:07] Cameron: it’s probably designed to be a, an early adopted developer research tool
[00:33:12] Cameron: while they keep working on it over the next couple of generations before it becomes more accessible, more useful, uh, cheaper for everybody.
[00:33:21] Cameron: But it seems to be a step in the right direction for fully immersive computing.
[00:33:26] Steve: Um, yeah, I haven’t tried it. Uh, the one thing that was interesting, I did wonder how many of the people, you know, skateboarding down New York and on the subway using it, was set up, you know, as, uh, some guerrilla marketing by, uh, the agency from, from Apple. I mean, I said the same images go round and round
[00:33:44] Steve: in circles.
[00:33:45] Steve: Look, I, I, I liked
[00:33:49] Steve: some of the feedback I heard about, you know, watching movies and that immersive experience seemed really extraordinary. I don’t know that me having goggles on my head and having a hundred screens open is anything that feels like what I’d want to use, because if it’s anything like my browser now, which has 646,000 tabs open, the last thing I need is more tabs open in different screens all around me.
[00:34:11] Steve: I’ll just be in a worse state than I’m in at this point. So, I, I dunno that I wanna spend three and a half grand us to, to have. The dismay that is my browser tabs just extended all around me and wearing something on my head to remind me how unorganized I have. I dunno if I want that. I mean, that’s kind of where I get to, but until I’ve tried it, I really, I really can’t make a comment.
[00:34:34] Steve: One thing really quickly on it is a lot of the naysayers are saying, oh, this is a terrible launch. This is no good. Gee, they’ve only only sold 200,000. They’ll only do 500,000. Let me just remind everyone of how powerful this organization is. Here’s a launch, which is getting dissed and people are saying is no good, and it’s got $700 million U.S in sales already.
[00:34:54] Steve: That is like bigger. That is bigger than the top 100 companies in the U.S, just with one crappy product according to everyone else. Just think that through for a moment.
[00:35:04] Cameron: Mm. All right. I wanna move on to the deep dive, Steve. Uh, my suggestion for this week is to talk about the breakthrough medicine stories that just seem to be coming faster and faster now. Just two of the ones that have hit my inbox in the last week or two, uh, uh, uh, uh, you know, so full of potential and look. Like that said, there’s always a lot of breakthrough medicine stories that we get that never see the light of day. There’s always gonna be a cure for cancer, or there’s gonna be this, or there’s gonna be that. We know that the research process is long and hard and that, you know, the PR departments for universities or publicly listed research firms are always trying to get these stories out to raise more funding or to pass legislation or whatever it is.
[00:35:53] Cameron: That said, these stories seem to be of a different caliber. This one from new scientist, thirty-First of January twenty-twenty-four. CRISPR gene therapy seems to cure dangerous inflammatory condition. Nine out of 10 people who received a new version of a CRISPR-based treatment for a potentially life-threatening inflammatory condition seemed to have been cured.
[00:36:19] Cameron: Nine people with a rare genetic condition that causes life-threatening inflammatory reactions appear to have been cured after taking part in the first trial of a new version of a CRISPR-based gene therapy. The condition called hereditary angioedema causes people to have sudden episodes of tissue swelling that affects body parts such as the face or throat, similar to aspects of an allergic reaction, although they can’t be treated with anti-allergy medicines.
[00:36:47] Cameron: 10 people who had the one-off gene treatment that has given directly into the body saw their number of swelling attacks fall by ninety-five percent in the first six months as the therapy took
[00:36:58] Cameron: effect. Since then, all but one have had no further episodes
[00:37:03] Cameron: for at least a further year while one person who had the lowest dose of the treatment had one mild attack.
[00:37:10] Cameron: This is potentially a cure, said Padmalal Gurugama at Cambridge University Hospitals in the uk who worked on the
[00:37:18] Cameron: new approach. So.
[00:37:27] Steve: I’ve heard stories and I’ve read some stuff and listened to podcasts on experts in that realm. Talk about CRISPR where they’re like, you can turn a cat into a dog. You can make someone six foot tall who was five
[00:37:39] Steve: foot three, like, you know, Cam. We can make your hair a different color. I don’t know if people who are listening know, but Cam’s just silver foxing it to
[00:37:47] Cameron: Why would I want it to be a different color? It’s
[00:37:49] Steve: Exactly, exactly. I would
[00:37:52] Cameron: you know, boss level hair color,
[00:37:54] Steve: boss level gray. Yeah, it is boss level. I hate my gray hair. But wh what I’m interested to know, and I’ve never seen this, is how do they do the gene therapy? And I think the listeners would know that CRISPR is like gene editing, where they come in and they cut out a, a gene, which is gonna cause problems.
[00:38:12] Steve: H how does the therapy get done? Like, I don’t
[00:38:15] Steve: actually know the answer to that. Please tell me, you know,
[00:38:17] Steve: cam,
[00:38:18] Cameron: Look, I’ve gone down the CRISPR path. You know, it, it broke in 2012, and I’ve, I’ve read about it.
[00:38:26] Cameron: I’ve listened to stuff about it as I, I can’t remember the specific details, but essentially it’s, I, as I recall it, it’s a tool
[00:38:33] Cameron: that was developed, uh,
[00:38:34] Cameron: sorry, discovered existing in nature. It was, uh,
[00:38:39] Cameron: something that a, a virus or a bacteria already did, and we figured out how to
[00:38:43] Cameron: replicate that. It basically enables you to find a specific piece of DNA
[00:38:49] Cameron: inside of a cell and then go in and alter that
[00:38:53] Cameron: piece of DNA using.
[00:38:55] Steve: it out. Like cut and
[00:38:56] Cameron: Yeah, cut it out and replace it with something else. So, um,
[00:39:01] Cameron: and
[00:39:01] Steve: how do you get that back into the body, and then how does it replicate across the entire DNA map? I’m just, I just don’t know how they physically do the therapy in inverted commas. Like how do they physically do it? You know? You know when you have an operation, you cut the person open and you take out the old heart and you put in a heart transplant.
[00:39:19] Steve: How? How do they do it? I’m just, I don’t know. I’m so curious. I need to dig in.
[00:39:25] Cameron: Yeah, how it goes from, okay, we’re gonna go change some cells
[00:39:28] Cameron: to propagating throughout the body, uh, fast enough to cure a disease. Yeah, I dunno the answer to that, Steve, but, you know, uh, it, it’s an incredible
[00:39:39] Cameron: piece of technology. As I said, like it first broke around about 12 years ago, became, uh, quite hot.
[00:39:48] Cameron: Here we are little over a decade later, using it to apparently solve a fatal disease that some people
[00:39:57] Steve: And, and almost any disease that, you know, I’ve read stuff that says that once we get down the track of it, it could potentially cure any disease where it’s genetic in nature and diseases that you have a higher
[00:40:11] Steve: probability of getting because of a certain genome. I mean, we just think about technology curve jumps.
[00:40:16] Steve: This would have
[00:40:16] Steve: to be the biggest curve jump of all time to get around pharmaceutical and a whole lot of other industries. Just think about it, you know, we’re editing
[00:40:22] Steve: and pasting and cutting and pasting our bodies to, uh, get us to the Gattaca level of, uh, however we want to be. I’m, I’m thrilled. I can just be the person I’ve always wanted to be.
[00:40:32] Steve: Cam.
[00:40:33] Cameron: Me just wanna be
[00:40:35] Steve: That’s right. That’s, that’s right, that’s right. Boss level.
[00:40:39] Cameron: boss level hair colour. Yeah, and it’s, and it’s relatively cheap too, to roll to produce cures using CRISPR to, and to deploy it too, which is the other big thing. The other story I saw this, uh, actually broke, uh, in December, but I only just came across it the last couple of weeks. This is, uh, Moderna’s mRNA cancer vaccine.
[00:41:02] Cameron: Works even better than thought. Adding Moderna’s in-development cancer vaccine to a standard treatment for melanoma dramatically reduces cancer survivors risk of death or recurrence. According to newly shared trial data. The challenge to treat melanoma, the deadliest type of skin cancer doctors typically start by surgically removing as much of the cancer as possible.
[00:41:26] Cameron: They might then administer another treatment such as chemo or radiation therapy to kill any cancer cells. They missed. But of course we know that the cancers can come back, they reoccur, and that’s usually what gets you, when it comes back, there’s a fairly high risk of recurrence, even if you’ve had it cut out or you’ve had chemo. What this, uh, so people obviously are familiar with mRNA, primarily from the way the vaccines were developed extremely quickly. Uh, during covid. mRNA hadn’t really been used, uh, on scale for human application before the Covid vaccines were developed, and Moderna developed one of the first vaccines using mRNA.
[00:42:08] Cameron: So what they’ve, what they’re able to do in this trial is if you’ve had a particular form of cancer, they can take a cancer cell from your body and then build a specific vaccine for you based on the kind of cancer that you’ve got. Using the same mRNA approach that they used for covid. Now people may recall the, basically the way that works is they would take certain indicators in the Covid case from the spike protein, um, put that into your body in a way that it’s not dangerous to your body.
[00:42:49] Cameron: It’s not the virus they’re putting back in, it’s just certain indicators from it. And your body would then develop antibodies that would look for those, uh, indicators and then target any cell that has them. They’re doing the same thing in this, but with melanoma cancers. So you’ve already had a cut out and then you’re building, uh, basically, uh, they’re called neoantigens.
[00:43:13] Cameron: The vaccine works by instructing the body to make up to thirty-four neoantigens. These are proteins found only on the cancer cells, and Moderna personalises the vaccine for each recipient so that it carries instructions for the neoantigens on their cancer cells. The idea behind the vaccine is that by prompting the body to make these proteins, it could prepare the immune system to quickly identify and attack any new cancer cells, bearing them, preventing reoccurrence.
[00:43:41] Cameron: Now in twenty-Twenty-two, they reported that the combo therapy reduced high-risk patients risk of reoccurrence or death by forty-four per
[00:43:50] Cameron: cent, compared to just the standard treatment in the two
[00:43:55] Cameron: years after treatment. They’ve now announced that people who received both therapies were forty-nine per cent, less likely to experience recurrence or death a meeting of three years after the treatment compared to people in the Hetruda-only group.
[00:44:10] Cameron: So yeah.
[00:44:12] Steve: bespoke medicine, you know, based on MRI. That’s, that’s, I mean, that’s kind of the holy grail, isn’t it? Because what we’ve had along this time is one size fits all. And you even with different medicines that that, that I’ve had, you know, it’s
[00:44:24] Steve: like the side effects are different for different people.
[00:44:27] Steve: This, this idea of, you know, design a medicine
[00:44:30] Steve: designed for the person is absolutely extraordinary. And, you know, I think we, we’ve been so focused on digital lately and, and technology that is consumer and information based. It’s, it’s, it’s interesting and cool that some of the things that are just ticking along, uh, happening within the healthcare realm, you know, it probably doesn’t get enough attention really, does it?
[00:44:54] Cameron: Uh, yeah. No, I, well, it’s hard, like there’s so much noise in the space of health stuff that it’s hard to tell the real stuff from the fake stuff. But these two,
[00:45:04] Steve: yeah. Wellness. Wellness And vitamins. Wellness and vitamins.
[00:45:08] Cameron: I wanna talk, uh, when we get to the futures forecast a little bit about what these sorts of personalized medicine stories may mean for the future of humanity.
[00:45:16] Cameron: Before we get to that technology time warp, Steve.
[00:45:20] Steve: Oh, I love this one. Oh my God. When I saw this, I was so excited, so excited.
[00:45:25] Cameron: anniversary this week of the Super Bowl Macintosh ad, the 19 eighty-four ad. And, uh, you know, it was a big deal then. It’s still a big deal. Now, New York Times did a big,
[00:45:43] Cameron: uh, article on it which is a good read. Uh, you know, not only was it a big deal for launching the Macintosh, uh, the ad was directed by Ridley Scott, who I still haven’t forgiven for his Napoleon film.
[00:45:58] Cameron: But, um, you know, this was, this was, uh, one of the highlights of his career. Obviously famously it didn’t show the product because they didn’t have a product ready to show.
[00:46:12] Steve: Is that the reason? I didn’t know that. I knew that it didn’t show. it.
[00:46:15] Cameron: Yeah. It wasn’t, wasn’t really ready. And the, the story, according to the New York Times is the product was gonna come out in 19 Eighty-Four, but it wasn’t ready in January when the Super Bowl happens. And they wanted to, they all thought that, you know, the ad agency Chiat Day knew that a lot of people would be probably gonna use George Orwell’s 19 Eighty-Four to advertise things in 19 Eighty-Four.
[00:46:42] Cameron: So they wanted to
[00:46:44] Cameron: everyone to the punch.
[00:46:46] Steve: Yeah, yeah,
[00:46:47] Cameron: So they, they, uh, put this together even though the product wasn’t ready, they couldn’t demonstrate anything just to get, uh, uh, uh, out ahead of everybody else and steal their thunder. But the great, it’s a great story. I mean, okay, so it not only was a great ad, but it started the whole Super Bowl ad phenomenon.
[00:47:05] Cameron: It was the first really big Super Bowl ad.
[00:47:09] Steve: Yeah, I didn’t know that it was the first one. And, you know, the Super Bowl’s
[00:47:13] Steve: extraordinary now, I mean, for me, I look forward to it each year for the
[00:47:16] Steve: advertising because I give, I think it, it, it, it, um, puts a stake in the ground for the zeitgeist
[00:47:21] Steve: on, you know, which products are doing well during the
[00:47:23] Steve: dot-com boom, you know, every, every venture, capitalism,
[00:47:26] Steve: pets.com and everyone had an ad. it’s like, yeah, I’m making a Super Bowl ad too. It was like, why did I do this startup to make a Super Bowl ad? Look, the technology is irrelevant. I don’t care about this startup. All I care about is I’ve got enough money to make a Super Bowl ad and I’m gonna do it. And so, yeah, crypto was three years ago.
[00:47:44] Steve: It was all about the cryptos, it was FTX and all of that. So it really is. A moment that tells you where we are economically and socially, but I didn’t know that that was the first big one. One other, um, note, uh, about the 19 Eighty-Four ad that I thought was really cool was it was the ad was only ever shown once and that was the only time it ever got aired.
[00:48:05] Steve: Every other time anyone has ever seen it then has been in pop culture and media referring to it. And so it was incredibly cost effective that it’s, yeah, here we are, what, 40 years later and we’re still, um, talking about it. So
[00:48:18] Cameron: Do you know why it was only shown once?
[00:48:22] Steve: I don’t know why You better tell me. Ken
[00:48:25] Cameron: Here’s the story, um, from the New York Times, John Sculley was the CEO of Apple at the time. Famously,
[00:48:32] Steve: in a can.
[00:48:34] Cameron: what?
[00:48:35] Steve: Sugar in a can guy.
[00:48:36] Cameron: in a can. Yeah. Steve Jobs hired him by saying, do you wanna change the world or do you wanna sell sugar in a can? Or something like that.
[00:48:44] Steve: Such a good line. So
[00:48:46] Cameron: And then Sculley fired Steve Jobs. But anyway, before that happened, Sculley says, before the commercial ran, we had to take it to the board of directors.
[00:48:54] Cameron: The board sees the commercial, then there’s just dead silence in the boardroom. They turn and look at me and a board member says, you’re not really gonna run that thing, are you? Another guy says, as the cloning credits scrolled up, the chairman, Mike Makula put his head in his hands and kind of folded over the conference table.
[00:49:13] Cameron: Then slowly straightened up and proposed hiring a different ad agency. I. Ridley. Scott says, I made it. I thought it was pretty good, but I was thinking, really, they’re gonna run this on the Super Bowl and we dunno what it’s for. Another guy says, I had them do a theater test. We get back the results and it’s the worst business commercial that they’ve ever tested in terms of persuasiveness.
[00:49:36] Cameron: Scully says, the board said, we don’t think you should run it. Try to sell the time. So basically the board said they, they bought this time to show it and the board says you have to sell it. You can’t, you can’t run it. Get rid of the ad space time. Then Chiat Day, the agency told them to drag their feet when they were told to sell off the time on the Super Bowl, which they did until it was too late to sell it off.
[00:50:06] Cameron: No one could prepare an ad in time so that they had no option but to run, go through it and uh, run with it. And then it says. Every news show had clips of it. the commercial kept running and running for days after that, it ran for free over and
[00:50:22] Cameron: over again. the
[00:50:22] Cameron: value of the offshoot publicity is
[00:50:24] Cameron: what many advertisers see as the bigger benefit, really.
[00:50:27] Cameron: Scott says, I think the Super Bowl frenzy started there. Then it was about
[00:50:31] Cameron: a million dollars a minute. Now it’s about $7 million a minute. And then the New York Times says Actually, the
[00:50:37] Cameron: average cost for a Super Bowl ad this year is twice that 7 million for a
[00:50:41] Cameron: thirty-second spot.
[00:50:46] Steve: One of the things that it, that it did was the cut through. I mean, and this is one of the great things in advertising. I mean, two things come to mind. A lot of great marketing that we look back on historically was irrational and courageous at the time. And for every courageous one that succeeds, there’s ninety-nine with people living under bridges saying, yeah, I invented new Coke.
[00:51:05] Steve: Like from that scene in the Simpsons, you know, it’s, uh, you know, courage gets you there, but ninety-nine percent of those courageous moves end up being disasters. So we’ve gotta keep that in mind. But the cut through was extraordinary. And, and the ad that stood out the most for me during the Super Bowl, I haven’t seen all of them, but I’ve seen highlight reels, was, um, I, I thought that what Kanye did was incredibly interesting.
[00:51:30] Steve: And, and I know that he’s, uh. He’s an, a weird and interesting person, has done a lot of kind of crazy stuff lately, but I dunno if you saw it. He had a, uh, an ad that he put on the Super Bowl, which was him like driving in a car. Holding up the phone, talking to it. And the first thing he says is, Hey, it’s Ye here.
[00:51:51] Steve: Um, here’s my Super Bowl ad. Look, I haven’t spent any money on production because I spent all the money on buying the ad. So he spent his 7 million bucks to get his 30 seconds. And it is literally him holding up a phone. I don’t know if this is him being ill and a not well person or some sort of genius, uh, uh, Kanye marketing stunt, but he said, so the ad is go to Yeezy’s dot com, we’ve got a whole lot of shoes and things for sale there.
[00:52:17] Steve: It’s gonna come up on the screen, and then it’s got like this little weird Yeezy’s dot com. It’s like an infomercial from 1987 at three in the morning. And he just holds up the phone and that’s the ad. And my TikTok was filled with it. I mean, so he has had a lot of, let’s call em reruns. Man. You gotta look at this thing in my freaking mind that that won the Super Bowl, you know, low production values.
[00:52:39] Steve: It was the antithesis cut through again in the opposite direction.
[00:52:43] Cameron: it’s clever. Well, I mean, the thing I, I was talking to Chrissy about this being the 40th anniversary in the car on the way to come through the other day, and then Fox started asking me what the Macintosh was and I explained and said, we got home that night and he wanted to see the ads. So I showed him the ad and then he wanted to see videos about the Macintosh, and we went on this whole YouTube rabbit hole of the original Macintosh.
[00:53:08] Cameron: And then he built one out of cardboard. He wanted to make his own Macintosh and he sort of became really deeply interested in the history of the Macintosh and the history of computing. So that was kind of cool. The ad is still inspiring kids generations later into understanding more about technology.
[00:53:28] Cameron: And I was talking about why it was such a big deal when it came out and I never had one. And you know, none of my, I didn’t even know anybody who had a Macintosh
[00:53:38] Cameron: in
[00:53:38] Steve: No one had, no one had computers back then, but YouTube, it’s a, it’s a good point you raised with
[00:53:42] Steve: Fox, is that. I, I really love the idea, and I do this with my kids a lot where I say, and I’ve got
[00:53:48] Steve: this parlance. I say, let’s go to
[00:53:49] Steve: The tapes, guys. That’s what I tell ’em. It’s, we didn’t go to the internet Google.
[00:53:52] Steve: I say,
[00:53:52] Steve: let’s go to the tapes. It’s, I want to get that out In modern
[00:53:55] Steve: society listeners, whenever you wanna show someone something, don’t
[00:53:58] Steve: say, let’s go to the internet, let’s Google it Say let’s go to the tapes. And we go on these little historical journeys with a whole lot of
[00:54:03] Steve: things and it, and it’s an incredible resource to take you back to what was at a certain time, whether it’s tv, advertising, history, politics, an incredible resource.
[00:54:12] Steve: And it can really be enlightening for kids if you take that time to give them some historical perspective. Because I think that having story arcs of where we’ve come to and where we are really creates an educational ballast. I really like that.
[00:54:25] Cameron: I agree. Alright, I know you got a heart out in a few minutes. I just wanted to finish on a bit of a forecast about transhumanism, Steve, reading these medical stories and thinking about it, life extension, um, you know, what happens in the next 20 years when we have medicines that can solve or cure almost any disease, aging, cancer, all the big killers, Alzheimer’s, uh, et cetera, et cetera.
[00:54:56] Cameron: Um, and people can live, uh, to a hundred, a hundred and twenty-five, 150 healthy, productive lives. When we have, you know, as we talked about in our last episode, potentially an economy that’s run by AI and robotics and nanofabs where you don’t need to work. Where you, you know, there’s this whole idea of, you know, reinventing what it means.
[00:55:26] Cameron: The human existence. Human existence has obviously been, what’s the old saying about brutish short and something, um, what happens when we don’t have to worry about cancer? We don’t have to worry about Alzheimer’s. We don’t have to worry about, you know, we’re probably not going to stop people from dying from an embolism or, or getting hit by a truck, or committing suicide or getting shot or those sorts of things anytime soon. But those sort of big diseases, the things that really
[00:56:02] Cameron: take on, take most of our friends and family in the West, you know, it’s different if you’re living in Gaza at the moment and you’re getting bombed by Israel. But if you, you know, those of us in the west that that worry about have, you know, friends and family and ourselves dying of cancer, et cetera, et cetera, when that goes away and we don’t have to worry about those things anymore, what does it mean to be human in that kind of a future?
[00:56:24] Cameron: Steve, how do, how do we reinvent ourselves? I.
[00:56:27] Steve: I mean, well this is the first one. It’s like it is like you can renovate a car where you can replace every part or
[00:56:35] Steve: you know. Buff every part so that it
[00:56:38] Steve: doesn’t get ill. I mean, are you still the same person? Does transhumanism mean that we’re just extending our lives?
[00:56:46] Steve: Or are we sort of transitioning ourselves into another,
[00:56:50] Steve: not portal, sort of another, what,
[00:56:52] Steve: what’s the word for it?
[00:56:52] Steve: Like another body of sorts. What? I just wonder whether or not we’re going to stay in the same form or whether we kind of become cloud beings to some extent. You know, that that’s, that’s what I wonder about. But if it is that everything can be made and done for us, I, I think it’s, it’s gonna cause a lot of psychological problems.
[00:57:14] Steve: I really do. I, I just reckon that there was something about the simplicity of life where, you know, you’re so busy doing and keeping the lights on, that it just has this purity to it. Like that, that pain of just being able to keep on, keeping on with life and, uh, being able to feed your family and get it done.
[00:57:34] Steve: You haven’t got. Time to become neurotic. And I feel like society is so neurotic because so many things are done for us and our heads just filled with so much stuff. I, I don’t know. Is it, is it, do you want to live forever? I mean, do, do people wanna live forever? don’t know.
[00:57:50] Cameron: I mean, I don’t know about forever, but.
[00:57:52] Steve: I don’t want to die. I mean, of course
[00:57:54] Cameron: that’s a model of time. But I’m here to tell you, there’s something
[00:57:57] Cameron: else. The Afterworld, the world of never-ending happiness. You can always see the Sun Day all night, but where do you come up that shrink at Beverly. Hills, you know the one doctor.
[00:58:04] Cameron: Everything will be all
[00:58:05] Cameron: right. Instead of asking him how much of your time is left, ask him how much of your mind, baby, because of this
[00:58:10] Cameron: life
[00:58:11] Cameron: things are much harder than the afterworld. This life, you’re on your own. Bum-bum, Bum-bum. And if the elevator tries to bring you down, go
[00:58:21] Cameron: crazy. Punch the higher floor
[00:58:25] Steve: Can you send me that song? I, what is you,
[00:58:29] Cameron: It’s the first track off of Purple Rain, man. Let’s go crazy. The opening
[00:58:33] Cameron: monologue.
[00:58:34] Steve: you go. I love
[00:58:35] Cameron: 1984, baby. You had to be
[00:58:36] Steve: look, I think, I think, I think the medical stuff and transhumanism and life extension is gonna be the really next big thing for sure. Especially that robotics is gonna take off a lot of the production, you know, nanophobes, all of that. It’s funny ’cause my wife works
[00:58:51] Steve: for, um, a wellness company and they’re talking about not just life extension, but they’re talking about health span, you know, not just lifespan.
[00:58:58] Steve: And I
[00:58:59] Steve: think that’s the key thing. I, if you just live a lot longer and you’re kind of infirmed for the next
[00:59:04] Steve: 30 years, I I don’t
[00:59:05] Steve: think anyone wants that. But if your health span is increasing, you can still do karate and go surfing. Like seriously. It, you just, that’s an, no one is gonna say no to
[00:59:14] Cameron: If you’ve still got your marbles and you’ve still got your health and your mobility, then what? What do you do with the next a hundred years of your life? What is your
[00:59:23] Steve: Maslov’s heart self-actualization. You do the things that. Turn you on, you, you do the things that you’re interested in. You know, if I was healthy, I would just want to get as good at languages and surfing and, and do all of those things that I love and grow vegetables and cook and just all of those really enjoyable things, just all the time, you know?
[00:59:43] Steve: And
[00:59:43] Steve: you might have time to.
[00:59:44] Cameron: I was saying to Tony on our show, the other that I’ve just gone back into Gogol and Dostoevsky and there’s all these classic books that I’m still
[00:59:52] Cameron: in my fifties working my way through. I haven’t got to yet. There’s probably hundreds of books that I, I really wanna read that I, I don’t have, I
[00:59:58] Cameron: haven’t had time to get to yet.
[01:00:00] Cameron: Films to see, music to listen to, you know? All right. This has, uh, been Futuristic
[01:00:07] Cameron: episode 21. Thank you Steve Sammartino. Follow us on
[01:00:10] Cameron: Twitter TikTok, um, MySpace, GeoCities, And uh, Friendster.
[01:00:20] Steve: Yep. And we’ll, Sam, we’ll see. We’ll be logging into the, well tonight at 1130.
[01:00:23] Cameron: Wow.
[01:00:25] Steve: The. Well.
[01:00:26] Cameron: If you dunno where to find us, look us up on Archie and Veronica and, uh,
[01:00:32] Cameron: what was the other search engines back in the day?
[01:00:34] Cameron: Um,
[01:00:36] Steve: Answers With a Z.
[01:00:37] Cameron: ANZs, that was the Australian
[01:00:39] Steve: answers. That was the Australian one. That was the, uh, Aussie male
[01:00:43] Cameron: I worked at Ozymail
[01:00:44] Cameron: when we had answers. Yeah,
[01:00:45] Cameron: GeoCities. Yeah.
[01:00:47] Steve: DOCities and Booksmart.
[01:00:49] Cameron: Yeah. Yeah. Oh. Um, what was the one named after? Uh,
[01:00:53] Cameron: Butler. Um, ask Jeeves. Look
[01:00:57] Cameron: us
[01:00:57] Steve: I feel like ChatGBT missed a great opportunity to call it Ask Jeeves, Finally, there would’ve been a Jeeves.
[01:01:02] Cameron: Sam keeps talking about how it’s the worst brand ever. Like it’s a really, they didn’t plan for it to be a product and now they’re stuck with it.
[01:01:09] Cameron: It’s like really bad brand
[01:01:12] Steve: that, that’s good, right? I mean, a brand here, let me just remind everyone. A brand is, it’s a cognitive shortcut from which to make informed decisions.
[01:01:19] Cameron: Nice. With that bit
[01:01:21] Steve: Copyright Steve Tino, 2 0 2 4.
[01:01:24] Cameron: Talk to you next time buddy.
[01:01:26] Steve: See you mate.