When Kurt Russell meets Donald Trump, and AI wins? January 24, 2025

23:27And that's the other problem. It's more ideological and practical, right? You and I would say, hey, maybe we can save a bit of money here by doing this and the other as opposed to government must be smaller, I must find savings. It's the right sizedness. But for all that said, the one thing he's absolutely right about, and we've talked about a lot, is he's absolutely saying, we need to deal with the deficit. We need to deal with debt. He's one of the very few people, and I will say, hopefully a positive impact on Donald Trump in that sense. I don't think I've ever heard Trump talk about the debt and deficit.

23:53I may have missed it. I don't follow politics slavishly, but the headlines don't cover it. That's probably because it's not sensational. But either way, the fact that someone with hopefully the year of the president is talking about that stuff has got to be good for America. And again, by definition, even if Trump is a bit America first and stuff the rest of you, a healthy American economy is good either way for the rest of us. If the American economy was to fail meaningfully, devaluations or non-payments, defaults, whatever, Those are bad for the rest of us as well. So if there is some good to come from this, saving your taxpayer from spending that I don't have to spend, that'd be really good.

24:31But dealing with that debt deficit, I think, is a really important one. Yes. However, they unfortunately are only focused on one side of the ledger. So we recognise the problem, right? And it's like, okay, so are we spending beyond our means? How do we fix it? We cut costs. They're like, well, yeah, that'll do it. um although there is the practical reality of how much you can actually cut those so which you've already mentioned the other side i'm talking about of course is well you could increase your revenue you know another way of saying you can increase taxes yes and that is an um anathema to from from a republican standpoint so so then if anything they're going to cut taxes further and probably for the rich so that's that's why i'm more i i agree i like i like that that's in being discussed, but I just find it is, I very much have the view that when push comes to shove, there's nothing they can do about it because cuts on their own won't do it.

25:27And there is no way on God's green earth that they are increasing taxes. It's certainly not anywhere near the extent necessary to match the shortfall. So, to borrow a well-used phrase, nothing stops this train. Shout out to Lynn Alden, macro analyst I'm quite fond of. But yeah, she's absolutely right. It's just maths at the end of the day. And there is, despite the rhetoric that the Trump administration is going to go in the same direction as all previous governments. And because they're kind of locked into things, it's just sort of like, you imagine an American system which says we're not paying support to the men and women of the armed services, you know, like that, no way, you know, or that we're actually going to cut back on military spending.

26:12Like, no, especially with the rise of geopolitical tensions, you know. You go through the line items there because, again, at a high level, we need to do this. Yes, 100 % agree. Here's the budget. What would you cut? What would you raise? Now, that is a much more difficult conversation. And the reality is we can talk about it all day long. Our listeners can debate it. And it's a worthwhile, fruitful conversation. But the reality is that every single, no matter which direction you sort of go in, there are compromises and there are trade-offs and there's going to be some group that's upset. So, the real question here is if you really want to fix it, who do we want to, I won't swear, who do we want to annoy?

26:53Thank you. Who do we, right. Someone has to be worse off. Correct. That's how it works. Either because they're being taxed more or they're receiving less services. And that won't, you know, again, it's not a black and white binary thing. There'll be some people who sort of win in some areas, losing others. But, you know, at large, someone has to be worse off. And that politically is always hard to do. And, you know, the left will sort of try and make sure that it's the, quote, unquote, the rich that get punished. The right will sort of favor the other sort of side of things, generally speaking.

27:26It's a pretty broad general statement. But my only point is that, like, hard decisions have to be made. Politicians aren't good at making hard decisions because hard decisions lose votes. I don't remember the last time we had a policy with losers. That's the trouble. Yeah, right. I'm trying to think about it now, and I honestly – I mean, there would have been some at some point somewhere. I don't have real serious policies. You know, there'll be red tape. People say, I'm losing. I'm going to do more paperwork or something. But, you know, genuine cost-me-money decisions, services cut, taxes raised.

27:55I just can't remember the last time it happened. And I think that's the only thing about every economist has said, we're going to have to raise taxes at some point, given the coming burden of the services that we say we want. We're either going to choose to have fewer services or raise taxes. There is no third way. There might be, again, we're going to take some dojish cuts to government and take some things out. But again, guess what? That means you're just losing the service, that you don't cut those costs again, as we just said, without saying to somebody, you're getting less of what you used to get because I've got to save some cash here.

28:21We don't do it. It hasn't been done for a million years. Motley Fool Money. For more, subscribe to the free newsletter at fool.com.au forward slash listener.

28:34Hey, Project Stargate. Yes, let's talk about Stargate. Yes, yes, yes. So not the TV show with Richard Dean Anderson, although that has a lot going for it. That's MacGyver for those of us who are a bit older. Kurt Russell in the original movie. It was Kurt Russell in the movie, that's right, yes. Not a bad show either, I quite liked it. So this is not the TV show, although I suspect someone somewhere, probably Elon has decided the name is a good idea because it's kind of fun and sci-fi. But Project Stargate is the name being given by President Trump to a$500 billion US dollar, so$800 billion Australian dollars, project to try and wrest back or take or grow or whatever you want to say, the AI and data center gravity from China to the US.

29:21It's the idea that basically more data centers, more cloud computing is going to be required. AI is a whole other layer of that, which is probably multiples of the size of what we already know as cloud computing and data storage. And Trump has stood on the steps of the White House with Sam Altman from OpenAI, the guys who do chat GPT, the guys from Oracle, others besides, basically unveiling this new program to try and win this data center war for the US. I mean, there's a couple of angles about this. The political one, there's the sheer size. I'm going to say, by the way, I think that number is almost certainly rubbish.

29:50It's one of those big numbers you put up, like Albo's 1.2 million homes. You can't pick a big number. It can sound impressive, but, you know, will it happen? If it does, it'll be by accident of, you know, commercial reality rather than whatever the government does or doesn't do it. That's not a slide on Trump. It would be the same if it was Biden or Harris or someone else. We'd say$500 billion, it's a nothing number. I don't think he's – just quickly in a joke, I don't think the government's paying anything of it. Yeah. So it's a real – that's an even better announcement you can make. Exactly.

30:17This is from SoftBank, Microsoft, OpenAI. They've raised a lot of money. Yeah. Yeah. So it's very easy to go, look, we're doing this and all these jobs are being created. It's like, oh, great. How much are you tipping in? Oh, nothing. I just made them stand on the steps with me because I told them I'd threaten them if they didn't do it. 100%. So I don't know, mate. One more very quick thing, too. I'll just mention that Elon was very instrumental in the establishment of OpenAI. Yes. Him and Sam have since split. Surprisingly. They don't get along that well. By the way, there's my prediction for the coming term.

30:51At some point, Elon and Trump are going to butt heads. That's it? Yeah, yeah, yeah. Guarantee it, right? Those egos, two egos cannot fit into the same room. So there's going to be that. But yeah, so Trump has got nothing to do with open AI. He's quite critical of it these days. Musk, you mean? Sorry, Musk. Yes, Musk, sorry. So I guess then to that point, apologies in co-opting a commercial decision that's already been made. It's no surprise, not new, and will happen again here and there, no matter who's in charge. I suppose there's a couple of questions, a couple of angles. One is just the AI thing itself.

31:27We've talked about it before, but I think whether the number is half or a quarter or a tenth of what they're talking about, it's still a huge amount of money. And the fact that these companies, commercial businesses are talking about investing anything like this sort of cash in these sort of ideas tells us exactly what they think the future is going to be. No one's putting that sort of money up unless they get a return on it, unless they believe this is the future. I think that's just notable for its own sake from an AI perspective. I guess the other one geopolitically does impact us is, again, I don't really know the depths of Trump's honest views about China and whether it's genuine kind of just bad guy over there, whether it's mass xenophobia, whether it just plays to the base or it's all of the above.

32:08But the idea of we want to do this rather than China doing it is going to be a refrain of the next four years. And that's just notable because it's not a case of we think this will work. It is a case of we want to beat China to this thing, which talks a little bit about his priorities and maybe what we'll see, speaking of tariffs, from the US administration in the next four years. Yes. So, a couple of things. Back of the napkin envelope. Back of the napkin envelope. Back of the napkin calculations. Coffee hasn't kicked in yet, mate. In inflation-adjusted terms, this project is something like 20 times bigger than the Manhattan Project.

32:45That's extraordinary. And if anyone saw Oppenheimer or, you know, read anything about that from history, it was a big deal. It's a big deal, you know, really was a huge effort in coordination. And that was something in inflation adjustment terms, I want to say it's about$30 billion, right? So, it's just sort of like to be that much bigger, you know, is just sort of like it shows you the scale of things. And again, I just emphasize inflation adjusted. That's in today's dollar term. So it is a massive investment, even if they don't end up spending at all. So that tells you very much about what at least the backers expect from this thing.

33:21Yeah, right. As you know, I'm a bit of a technophile. I'm just a sucker for this kind of stuff. I just jump on anything that just sort of has, you know, it's hard not to find it cool. So Sam Altman and others in the space are talking about, in fact, they're increasingly talking about AGI, which is artificial general intelligence, which is a big quantum leap from where we are now. But before that, this year sometime, their newer models will be PhD level across all domains. Yeah. You know, so at the moment, if you've got a subscription using GPT-40 or using Claude or one of the other major models, they're good, right?

34:04They're good. But they're kind of like a graduate, you know, or pre-graduate level intern. You know, they're not great. And the newer ones, because they measure these things against all kinds of benchmarks. You have a PhD level kind of capabilities here. So, they'll throw them at all these kind of tests that we put the experts through in sort of doing things. And it's kind of getting up to those levels. Now, that's really interesting for its own sake. It's going to have all kinds of implications. But that might be one of the ways that they muddle their way through some of these fiscal dilemmas.

34:38Because it's just the sheer productivity boost that something like that could do. The hard thing for me is that, again, being around capital markets my whole life, you know that if you want to attract funding and you're in a business that has huge capital expenditure requirements, hype is your ally. right so you don't if you're sam altman you don't come out and go oh gosh it's 20 years away and it's lots of hard work and we may or may not do it no one's going to fund you when you say oh yeah we're on the cusp of phd level and you know um uh agents here and oh by the way we're going to have artificial general intelligence and by the way once you have agi you have super intelligence because it'll just it'll it'll compound on itself so you just go can you just make a better version of yourself okay and it'll do that and then the newer version you go can you make a better version of yourself you have this recursive improvement from something that's 10x smarter than you than 100x you know it's like it's pretty wild right so again i don't know i don't know enough about computer science and all this stuff to really handicap that but i will note that it's interesting that when you when you survey the experts in the scene there is a very high um portion of people who are, no, no, this is not hype.

35:51And as I understand it, and I'd love if anyone listening out there is more informed on this than me, and there's almost certainly that's true. The real, what's fascinating about this technology, in a lot of technologies actually, is you have what Peter Thiel might call the zero to one moment. We didn't know how to do LLMs, large language models, now we do. And it's like that is sort of like the proof of concept or the fact that it can be done. you go out of theory into the lab, into the real world, and it works and you demonstrate it. That's the hard thing. Going from one to two to three to four is often just a question of scale.

36:32And some of the podcasts I've been listening to lately, the view is that in regard to So ramping up AI to the capability that we hope and believe is possible is not reliant on a new breakthrough in computer science or understanding. In other words, if we just do the techniques and approaches and principles that we already know how to do, but we do it at a larger scale, we will get there. Right, okay. And that is different. So, when you say we're going to spend half a trillion dollars on AI and hope that we make the necessary breakthroughs to make that hardware worthwhile is different from – maybe just compare it to physics.

37:19So, anyone who likes a good super collider, a cyclotron, and likes to smash subatomic particles together, that's a scale problem as well, right? So, they keep trying – the physicists keep trying to build – I told you about the coffee – build bigger and bigger particle accelerators. They're doing the same thing. They're just doing it at higher energy. And when you do it at higher energy, you get to sort of peer deeper into the atom. So, when they build these big projects here, it's not a question of, you know, we know that if we have more energy, we will get more out of it. It's just, we just need to do it.

37:54The only bottleneck is the capital and the time to do it. I guess I'm not making the point very eloquently here, but I think what is interesting about Project Stargate is it seems to be addressing that problem of, again, assuming what the experts say is true, is like, we can do this. We just need more compute. Right. So, and that's the word that you'll hear more and more of lately. It's compute, compute, compute. Not a new way of doing things. We just need more compute. This will give more machines to the problem. More machines to the problem. I just need more of it, right? And we've seen it play out just in traditional computing, right?

38:33It's just like as we get more and more cloud-based infrastructure, as the networks get better, we just, again, the internet is running off protocols developed decades ago, right? It's just that we've got more capacity that's there. And with capacity, extra capacity unlocks extra capability. And that is the point that I'm really struggling to make here is that, again, if and if, and I'm just relying on what others in the field have sort of said, if it is really just a scale problem, it may well – you could probably justify the spend very easily. It's like it's an insane amount of money. However, we're pretty confident that we will get a very good return on that.

39:18And think about the return that you would have as an organization, as an institution, as a government, to be able to spin up a million Einstein-level PhDs and throw them at any task. And they'll never sleep, and they'll never question, and they'll never strike. It's pretty hard to wrap your brain around. And as I say, there's a whole bunch of reasons why the coming years are going to be crazy, but that's definitely one of them. so i i think that's i i share your i share your optimism and your excitement about it some dystopian edges to it by the way you need to be careful but that's oh i have to deal with as we go i i'm left thinking mate about the value uh creation though and where where the value to whom and to what the value will accrete and i guess i'm thinking about so yeah for a long time i said next dc it's a data center operator in australia i said look these guys are glorified REITs real investment trusts.

40:10They happen to have servers. Yes, they're specialized, but the kind of multiples I was trading on was stupid in my view because I could, you could, they could, Amazon could, IBM could. It's a commodity. Right. Could just simply do it side by side. So where's the margin? The answer was, I still don't think there is one, by the way, but given the future of this stuff, it strikes me that demand is going to outstrip supply potentially for years or decades. Yeah. That's my view. Yeah. So we just interrupt. Well, really worth hammering that point. I'll admit I got it wrong because when I was looking at a lot of these data centers, it was just like, what's the special ability that you have here?

40:47It's like anyone can buy a warehouse. I'm not saying it's easy, but I'm also saying it's nothing to stop those with capacity and capital to do it. But what I think I may have missed, and maybe the jury's still out, is just that, as you rightly say, that the demand, Even though we're putting so much supply on, the demand is so vast and growing at such a rapid pace that it will continue to outstrip even the massive ramp up in supply. Sorry, mate, continue. I just wanted to get that point. Yeah, that's right. And that's kind of what I'm now wondering. I'm not sure, again, as you're not, that I am right.

41:21I'm not sure whether that's enough, what these guys can do as a result. You've said lots of times, you know, price is what we get out of that. So at some point, if there's not enough supply, demand is too high. That means that the limited suppliers can charge more. At some point, that's a retarding factor in the growth because you've got to get a return on the compute you're using. And that's the other thing, I guess. If I think about everything from we'll use more energy, we'll need more data centers. There'll be more AI compute. There will be more uses of it. I'm still not sure for all of that where I expect the returns to accrue.

41:59I mean, to humanity, right? to your 100 PhD Einsteins, as you say, is where there's real obvious benefit. But commercially, do you have an updated view on who wins? Because I'm still not sure where I put my investment dollar on that basis. Again, it's the airline airport thing. In hindsight, the airports were where you invest and not the airlines. This time around, other data says the airports, maybe I don't want to be too literal with the analogy, but maybe it's that. Maybe it's the energy providers, frankly, for the air energy. You just get a skyrocket. Maybe you want to be there. Although, again, more volume doesn't necessarily mean more profit.

42:36We know that with air travel. We know that with oil. I've said for a long time, my suspicion remains most of the value will accrue to the users in disparate and diverse ways rather than there being a large number of winners who say, I am the AI king. I will charge a squillion dollars for this thing. because if one AI model is slightly better and maybe it's better in its domain, the chance, again, think about competition. We talk about that all the time. This is a microcosm of all that stuff we've always talked about, which is will one AI be better than the others? Maybe. Is it better enough in enough domains to be worth that much more than the others?

43:12I don't know. Is it the chip makers? NVIDIA is making a squillion dollars right now. Do they keep that dominance for 20 years or does Andrew Chip come out and take out some of that business and take volume and more importantly, take margin because now there's competition. I really don't have a view other than I wouldn't be betting on an AI play right now. Maybe data centers are firming in my mind because of what we just said, but do you have a view? Well, I'll preface it by saying having been someone who's taken the time to look at other revolutionary technologies and what was said at the time, it's not going anywhere.

43:50Okay. Of course it is. Well, maybe. Of course it is. We'll see. All roads lead there. We know that by now. But whether it was electricity, the internet or something else, you know, when we look back in hindsight, the takes were so quaint. They were so laughable. And it's – but I'm also very cognizant that they – at the time, they weren't. It's easy in hindsight to go, you know, obviously electricity was a big deal. At the time, it was considered hugely dangerous, massively inefficient, super unreliable, expensive, all of the things. And it's like if you were in the 19th century when this was all coming about, you're very intelligent, well-read, you know, people with access and connections.

44:39We're making these takes that were, again, stupid. And so that's why I preface it because I feel as though whatever I say next in 10 years' time, I'll look like I'm just face-par. But that is the point. That is the point, right? Yeah. Yeah. So there's that. But having said that, chips seem like an interesting area, but you've made the point, well, okay, NVIDIA is winning there, but, you know, all of a sudden Intel comes out with a completely new architecture tomorrow. We're in the same one, just over the price. It's so hard to know. Yes. Or someone in their garage completely revolutionized it in a way that just no one saw coming, right?

45:19Energy, I agree. I mean, these things are super energy hungry. And I just think in general, as a civilization, we should be doing everything we can to maximize our energy production in a responsible, environmentally cognizant kind of way. I'm not saying we have coal-fired power plants everywhere, but energy is the base layer of civilization. We should want as much as we can, and we should strive to make that as clean and as sustainable as we can. And I've said this before, too. I really think those that benefit the most are those with proprietary data sets, because when intelligence is a commodity, let me clarify, when superintelligence is a commodity, in the fact that anyone can spin up a gazillion PhD-level agents, well, the only advantage is that, well, I just have better data than me.

46:08You might have a smart bot. I've got a smart bot. Yours is trying to infer and guess, and mine's relying on a whole bunch of database, you know, a massive database that only I've got access to. So that's my best guess. Even then, I wonder how, what the ROI is on that, you know? How much better does it make you to your competitor that your data is a little bit better, given the bots are probably equally capable? I think you're right, by the way, but this is where I struggle is like, you know, who's got the best safety record? Qantas. Okay, well, they can charge extra$10 for their tickets, but no more than that.

46:39So, yeah, they're better than Virgin, and that's important because Virgin's, you know, barely profitable most years. Qantas is a little bit more profitable a little bit more often. Yep. And, again, I know it's not the same thing, but maybe it is. I guess, you know, the degree to which the data set, how much value you can extract from that versus the cost of having to extract it, that's the thing I think. We know it's a great thing, but, as always, the price you pay matters a heap and the return you can actually get versus your competitor is hugely important. I think, by the way, let me be very clear.

47:12I think everyone will get massive benefits. So I'm not saying there is no ROI. I'm saying incrementally. I use it and you use it. If your outcomes are slightly better than mine, is that enough? In some cases, probably. If the winner takes all, maybe there's enough because you're just a little bit better and that accrues and compounds hugely. Others are like, well, you can make your widget for$0.09. cents i make mine for 10 cents because your your way is a little bit better we're both selling for a buck it's no different here and and that's where the i still i just and i'm i'm being deliberately vague and i know it's annoying because people like absolute answers but i can just you know trying to grab a solid answer that's really tough no it's the only look honestly mate it's the only rational standpoint you can make at this point for exactly the reasons i started off before with because again in in hindsight our takes will look anyone who's speaking with certainty is going to have mud on their face, you know, and there'll be a few people who just lucked onto the right answer, you know?

48:07So, yeah, I'm really cognizant of what, and just the thing I guess I want to mention on the pod is that people are right to be excited about AI, I think. Yes, absolutely. But you're going to be, Wall Street, Martin Place, they're going to package up all kinds of things to, you know, when the ducks quack feed them, that's the saying in financial markets, right? And so, people will get excited and there will be all this kind of AI ETFs and the rest of it. And I just, that's the thing I worry about. So, you rightly be excited about it, but just embrace some of that humility that Scott was referring to there in the sense that it is very, very hard to sort of predict the trajectory and the timing of these kinds of things.

48:50And just because something has an AI badge on it doesn't mean that it's going to have any extra capacity to add value, right? Again, you and I are old enough. We've lived through it. I know it's a hackneyed old example, but the internet was going to change everything. And just like you give any special credence to a company that has a website and an email address. That's right. You do. Walmart.com listed separately because it was a.com. For a while, there was Walmart and then Walmart spun off Walmart.com because people were paying stupid amounts of.com email addresses and businesses that had them.

49:21So hang on, see, this was just, you know, go on. Well, I say unbelievable with the benefit of hindsight. Again, at the time, very, very capable, intelligent capital allocators, you know, the corporate leaders of the biggest economy in the world were making these bets. So, it's not, you know, it's not, I'm just cognizant of that. The other thing that is, I guess, from a societal standpoint, which is interesting, is that the poor have always had one thing that they've been able to do, which is you can always sell your labor. In a world where we have brains in a box and we can put those brains in robots.

50:01And by the way, I know I've said this before too, but I don't know if you saw some of the stuff that came out of NVIDIA recently. They're very much gearing up for robotics. There's a lot of money being thrown at that. Check out FIGA as a company. Anyway, there's a whole bunch of them. Tesla's obviously doing it with their Optimus spot. They're super impressive and it's growing really fast. In that kind of world, if you don't own assets, and most people don't own assets of the 8 billion people on the planet, what left do you have to offer? Like you could have always said, okay, I'll clean that toilet.

50:35I'll do that. I have an opportunity to contribute and build up, you know, deliver value and get value in return. That's kind of – so, I don't know. Is UBI's universal basic income? I don't know. I don't. So, that's really interesting. The other interesting thing is in a world of abundance, and ultimately that's the bottom line here. It's like what's the point of having all this great capacity if it doesn't give us more stuff, right? I don't mean from a base materialistic sense, you know, that I need 10 iPads and then I'll be happy. You know, I'm not talking about that, but I'm just sitting in a world where these incredible brains and resources allow us to have much, much, much, much, much more for much, much less.

51:18And even the essentials, you know, it's sort of like, what do you do as an investor when everything is abundant? Scarcity is something that I think you want to think about there. and that's going to be a function of that is going to be finding the companies that have unlocked the value the key to value with this technology because maybe it does consolidate like the internet consolidated around a few very very very big players and they will be scarce by sort of virtue of their position maybe that's something that you want to sort of focus on as well so again I guess let me tie a bow on this from my perspective and I'll throw it back to you But I think a high-level thought is you're right to be excited, but stay grounded.

52:06Don't buy into hype. Don't feel as though you need to invest now or you'll miss the train forever. And be prepared to change your mind as things evolve because they will evolve in ways that you can't predict. It might be that OpenAI just comes out tomorrow and just says, yep, we've cracked it. And now we are the best at everything in the world because we're the only ones that have cracked it. We have a first-mover advantage. We've got AGI. And before we told anyone about it, we just set it upon itself, and it's just now it's 4 billion times smarter. And that's actually why we're really building this thing, and it's impossible for anyone to catch up.

52:43And that's a very different world, in which case you have, no, a dozen or two dozen different independent AI companies that all kind of crack it at the same time, and we have a healthy competition between the two, as opposed to one dominant superpower that is almost definitionally uncatchable because again when you have god in a box you know it's just like um chat gpt fine for me anyone who's very any anywhere near coming close to cracking the the code and what can we do to stifle them or buy them out or whatever you know what i mean you get it is the amount of power that this unleashes you know even if it's just on a through the the pure lens of just capitalistic competition.

53:23The advantage that that gives you, it is a world, I hope, where we do just in the same way that it's good that there are multiple competing powers that have similar technologies. I don't know if we want this to consolidate around a single entity. I think that's right. I'm very sure you're right. Because then we do live in a world where someone like Elon Musk is the grand emperor of planet Earth and has, you know, and I know there's a lot of people. I can feel you, audience. I can feel you rolling your eyes. No, that's the dystopian reality, mate. That's the dystopian – what's that reality? Potential reality.

54:00There is serious – and to be fair, again, I don't like Elon Musk much at all, but his issue with Sam Altman was rooted in the fact that Altman took chat GPT or OpenAI as the company away from being this not-for-profit helping humanity thing into commercializing and making money from it. And Musk was like, hey, that's not cool. You know, this could be used for good. Now, at the same time, Musk is happily, you know, running Twitter and doing other things and, you know, using his own wealth for reasons that he obviously believes are worthwhile and someone's truth and someone's good for humanity is not everyone else's necessarily.

54:36But you're right. I mean, the chance this runs away from us is not zero and that's kind of where we need to be concerned about it. Sam Altman himself is on record as kind of saying, and we're not really sure how this works and we're not really sure what happens. We should be concerned. Not as in we should be concerned to shut it down, but we should be being very careful and very aware of what comes next and how we manage it because, yeah, you start the chain reaction and step back. There are reasons to be worried and reasons to be very aware of that the good stuff is absolutely great, the bad stuff is not guaranteed not to happen.

55:11Yeah, there's a real good podcast I can recommend in regards to the Lex Friedman podcast where he speaks to the team at Anthropics. So they're one of the big rivals to OpenAI. And the guys there are very much focused on what's called the control problem. If everyone really wants to get into this, I can highly recommend a book called Superintelligence by Nick Bostrom. And if you don't want to plow through that, it's a bit dense. There's a lot of good talks he does on YouTube. He talks about this kind of thing. And I am heartened the fact that they have put such a priority on this thing. And he really makes the founder there, and I forget his name at the moment, but he really makes the point that it's not something you think about after the fact.

55:50Because once you've developed AGI, it's too late. You think, oh, I just created God. How do I control it? It's like, no, no, no, no. It's too late by the time that kind of thing happens. So they talk about all kinds of trigger points that like, okay, once we have this capacity, we need to enact these kinds of safety protocols. Once we pass this threshold, we have to do that. And it is, again, this feels like it's a long way from investing, but it is super important to sort of think about those things and not do it in a sort of like a giggly sci-fi eyes if that's ever going to happen and people will talk.

56:23You can already see Channel 7, you know, putting up Terminator 2 images up there and giving the really lowbrow take on all of this stuff. I like Channel 7, by the way. I love your work. So bringing out the best in us with you and the other commercial networks. But do you get what I'm getting at here in the sense that this stuff moves incredibly fast and it's hard to conceive or think of a historical parallel because we've never had anything that can think and let alone think in terms of itself and self-improve and all of this kind of stuff. So, I don't know, man. All I know is it's going to be wild.

57:10And there is, of course, too, and I just need to put this out here for the sake of myself more than anything, not to be a wet blanket, but there is absolutely a future path forward where we just hit some unforeseen bottleneck and we go through another 30-year AI winter. Right. In which case, oh, it turns out that the current level of large language models is about as good as we get. And we do need some kind of fundamental theoretical breakthrough before we can go to the next stage. And that might be decades and decades away. And it turns out that a lot of this investment goes up in smoke and never really gets a good ROI.

57:45So, you've got to be alert to that kind of possibility as well. So, always, and this is true with just not just in this area, but all kinds of investing. is you have to form an opinion, but you don't have to act. And the only time to act is when you have a high degree of conviction. And if you don't have a high degree of conviction, don't do a damn thing. And then recognise that things change, new information comes to light. So you don't have to make a decision now. Your decision might be, this is interesting, I'll keep an eye on it, and then as new information comes forward, I will incorporate that, I'll think on that, I'll see if I can form some kind of conviction, and then I'll just keep iterating on that process.

58:25And as we've said in previous, I guess the other week or so, it will mean that you miss out on some of the early and quite likely spectacular gains that has the potential to be unlocked here, but you'll still do insanely well and much better on a risk adjusted basis. If you just wait to sort of see that dominance sort of emerge and then compound on itself. I think for me, that's what I'm trying to do. As much as I'm very passionate on all of this, I didn't have a single AI direct-focused investment. Yeah, man. You know? And someone wrote in the other day with one of the Mailbag episodes on quantum computing, super psyched about that too, but I'm not touching it.

59:02It's too early. It's too early. But maybe next week when we chat, we're like, oh, turns out this happened. Now it's a much more certain proposition, and I reserve the right to sort of change my mind. I think it's the only sort of rational way forward. I like it. Matt, I should say, by the way, I assume the audio was as good for this as for me. You also have some cockatoos in the background flying past from the sound of it. Speaking of Amazon, a delivery came and I ran out the front. I want to make sure you wouldn't run away without me signing it. I'm heading back inside. At least there's cockatoos this time, not cicadas.

59:36I hear the cocky screeching in the background. It was fantastic. I just thought I'd call out that for listeners who were thinking, particularly our international listeners. G'day if you're listening from somewhere else. That was in the background of cockatoos screeching as it flew past. They are very, very cool animals, but they're loud. Well, are they? Oh, yeah. Because I used to think so, but they love to destroy. It's just buggers. They're like, oh, here's something I can destroy. Oh, there he is, putting a plan in the ground. I'm going to eat that. And what do I say, eat? No, just destroy.

1:00:07Yeah, that's right. Yeah, just break for the sake of it. Yes, they also eat woodwork. So we have some chunks out of some window panes and part of our balcony that cockies have helped themselves to. Hey, let's move on. We've spent a long time talking about those things. I do want to bring it home just for one topic, mate, which is STAR. We talked about that before. And I just – it's notable, I think – well, again, speaking of hindsight, it won't take 10 years, but it'll be three months until we can look back and think about what we said and thought and did in January 2025 because STAR's got less than three months' cash left.

1:00:41And one way or the other, they're going to have to find themselves a path through this. not going to add much more to it other than to say they came out during the week and said there's no guarantee we are going to be a going concern. Now that phrase going concern doesn't mean a whole lot to non-financial types. It's the phrase the auditors use which basically means it'll be able to pay its bills. We're going broke. Right? So Star Effectively and they've been, I'll give them their credit mate and one brick bat. They've been pretty clear most of the time through this about exactly what's going on.

1:01:10You know when if and when Star implodes no one can reasonably say, oh, that was a surprise. We didn't know it. We weren't kept informed. Now, maybe it was self-interest and self-preservation by the directors, and if that's the case, that's fine by me, because incentives matter, right? We see that a lot. But they've done it as well or better than I think any other company I can think of, mate, that subsequently has gone broke. Again, some may not, so let me be really clear. It's been a decision for a handout. It's my cynical take on it. But at least they're doing that, right? They're doing it in a public way where shareholders can't claim to have not been informed.

1:01:39Think about any bankruptcies over the last 10 years. And yeah, the business is like, oh, is it bad? Is it bad? Is it bad? It's like, oops, it's in administration already. Mosaic brands, right? Was it always struggling? Yes. Was it closing stores? Yes. And all of a sudden one day it was like, ah, so we've called the administrators. And it's like, well, I'm not surprised, but, you know, and again, I'm not just, to be really clear, I'm not saying Mosaic did anything wrong at all. If they have, the court will, you know, someone will start a court case, but then the court can find what it finds. But Judge Starr has been, even if it is for, again, as other incentives, right, getting some cash, Shareholders cannot be anything but happy with what they've been told by Star, with one exception, which is directors have taken advantage of what they call the safe harbour provisions, and it basically allows them to continue operating when they are at risk of insolvency by basically coming clean with all the lenders and like, hey, here's what's going on, guys.

1:02:27And it gives the lenders more control and say in what the company then does, and it keeps the directors clear. Now, they hadn't disclosed that before the AFR reported it, and I thought that was pretty uncool. I'm sure it's legal. I'm sure they've followed the law perfectly. I'm sure the lawyers are making a fortune at a start. So, again, I'm not claiming I was doing anything illegal or improper. But in terms of absolute disclosure, I would have liked to – I'm not a shareholder – but if I was, to have known they were giving the lenders more control as part of that process. Some people have said, well, shareholder's still better off because if lenders are happy, then there's more time and space to get out of the funk, and that's probably true.

1:03:04But as a shareholder, I would have expected a company to let me know out of respect if nothing else. I think the law should be changed too, by the way. They should be forced to disclose it. But that's the only brick bat. Otherwise, bouquet is all around for those guys. But, yeah, when you say we may not be a going concern, and they've been increasingly clear and increasingly frank, frankly, with what they've said, which is kind of, yeah, we're in trouble. Oh, we're running out of cash. We may not be a going concern that the language is getting starker, and it should be, but it does go to show you how close they are.

1:03:37It's also made, I think, just a quick lesson in carefully – I think it's a gambling company, so it's appropriate. People, even on Twitter, after this announcement, someone said to me, oh, I'm going to buy tomorrow morning because the price has fallen, I'm going to make a fortune. And in one parallel universe, the shares went back to$0.40 from what was$0.15 or$0.20. And they say, see, I told you I made a fortune. I do this all the time. I'm really good at this. I'm going to make money. And they're not necessarily wrong, although this time around, they did that and the shares kept falling. And the only reason I point out is not for any shard or any blame or anything else.

1:04:10Just be careful with those bets that seem like they're smart or seem like they have to pay off. There's people who say, oh, I always buy the day after a bad announcement because shares always jump back up or, you know, I never do that or I always do this. Those things, there's no rule that says they have to happen and if there was, everyone would do it and the opportunity would disappear. So just be careful when you think you're getting a bargain. As we said many, many times, something's down 90 % with once that 80 % then halved from there. So it can happen. It does happen. So just be a little bit careful with, if you want to use your TAB money and punt it on star shares instead, knock yourself out.

1:04:44But if you're an investor and you should be, just don't, my suggestion, I can't give you advice, my suggestion is don't try and play funny buggers. Don't try and get the, you know, the gotcha kind of bounce back, the dead cat bounce. Sometimes it does, sometimes it doesn't. The odds are probably not in your favour, I wouldn't have thought. Well, yeah. And you've got to remember too that, you know, there are a lot of entities out there that have much better access, much better resources, much better training, if that's the right word, in all of this. And they're all trying to play some of these, as you say, silly bugger games, but you're not only playing a silly game, you're playing a silly game with an unfair advantage where there's others playing that same silly game, but they just tend to be much better resourced and connected than you are.

1:05:29Yes, exactly. Yes, you've got the internet. Well, they just had a meeting with one of the board members, right? Like it's just, wait, that's not fair. No, it's not. And it's egregious. But again, back in the real world, that's kind of stuff that sort of happens. So, yeah, be mindful of that. What else to say? Look, I would say it's easier for these guys to sort of be more frank with their shareholders, as you rightly pointed out last week, because they weren't the ones that drove it into the ground. When we say easier, we don't think easier is in it's more possible. It's just ego-wise. It's just our psyches allow us to do it when it's someone else's fault.

1:06:07It's a different CEO. So you can go, oh, look at that other bloke did it, you know? And it's just much easier to do that. I've come across that a lot of times. But I will say this, you can never know, right? You can never know for sure. But a lot of these things, there are signs along the way. And they're all heuristics and there's no guarantees with any of this kind of stuff. But, you know, there is something to be said as an investor when there's a lot of smoke. There's not often, I mean, there's not always fire, but there was a lot of smoke going around this entity for a long time, right? In terms of some of the accusations and some of the inquiries, you know, maybe it was all a big witch hunt and there was a big conspiracy against them, or maybe there was some truth to it.

1:06:53And again, at the time, management, the then management, was no, no, no, no, we're absolutely fine. No, don't worry about it. And even when you were more analytical with some of the numbers and stuff, you could see it in there. It's just like, gosh, there's not a lot of room. I mean, if you guys have a couple of bad periods here, you're going to tip into loss. And my gosh, look at the amount of debt that you've got. So, you've got a lot of tools at your disposal as an investor. Again, you can never know. And people, again, much smarter than me and better resource than me can probably tease it out much quicker and sooner than I can.

1:07:33But, you know, when you're trying to flee the Titanic, you don't have to be the first person on the lifeboat, right? You just have to be on at least the last lifeboat. and and and i i i often not often but what i do get to points where it's not like anything is obvious but when there's enough red flags it becomes at least to me a point in which case look i'm it might be the worst absolute possible time to sell but i'm certainly not going to not sell just for the for the purpose of what you might call ego preservation because we all like to go oh well it's not a loss until i sell it's really cheap now oh they're going to turn it around.

1:08:10So you'll make all kinds of excuses on that. And when it ever comes to the, I always like to frame it this way. If you wouldn't buy today, you shouldn't hold today. Now that's not true. You know, there are exceptions to that, but as a general rule, it's a pretty good one because when you think about it, the market doesn't know what you paid or when you bought your shares. It doesn't matter. And the fact is Star City is worth, what'd you say? 20 cents or Whatever it is now. It's 11 cents, I think, when we're recording this. I'll check, but yeah. Okay. But that's the point, right? Well, that's the point, right?

1:08:44Yeah. You thought when it went from 20 to 15, oh, 15 is cheap. I'll buy something. When it goes to 20, I'll make money. And it goes to 11. It's like, oh, okay. Well, I thought I was going to make a fortune. It's a bet. It's a gamble. It's a speculation. Again, use your tab money. Use your pokies money if you want to for that sort of stuff. But don't use your investing money for it when you're just literally hoping. It's not all this hope that maybe the share price might bounce. Absolutely. And my point is, if you are an existing shareholder, it's very easy to sort of say, well, it's down so much, so what difference does it make at this point?

1:09:13It's like, well, even if it's like$100, it's$100. Lose$100, yeah, exactly. Why? Why lose it for the sake of pulling out your smartphone and pressing three buttons? Like, it just makes no sense. Or to sort of say, gosh, I mean, again, ask yourself, if you didn't have the shares today and you were looking at Star City and you go, God, I wouldn't touch it with a barge pole. Well, then you shouldn't hold it either. Again, regardless of when you pay for it. Even if you're in profit, if you wouldn't buy today, you shouldn't hold today. I just really want to make that point too because generally speaking, a lot of money has been lost just by people failing to acknowledge a mistake has been made.

1:09:55And mistakes are made all the time from every one of us, all the time in this game. And if you're going to like deny the obvious, the reality of the situation, you're just going to make a bad situation worse. Pick yourself up, dust yourself off, and put your money somewhere else. I'm not saying it's easy emotionally, but it's absolutely the right thing to do. I'm not going to tell any shareholders what they should do with their star city shares, other than if you wouldn't buy today, you shouldn't hold today. Yeah, and the reason, by the way, people say, oh, the money's gone, the money's gone, whatever, the buy-hold thing.

1:10:24The mental exercise is if your entire portfolio was liquidated overnight, what would you buy Monday morning? When the market opens Monday morning, Yes. Would you buy the same companies? And the same for every, not just start, every asset you own. Every company. Right. Would I buy this in the same proportion at the same price today? If I had, I've got a$1 ,000 portfolio, it's been liquidated accidentally. Comsec called me and say, look, really sorry, we'll refund your brokerage, but we actually sold everything. What do you want us to buy back for you? Would you actually say, I want exactly 1 ,000 star shares and 200 BHP shares and 400 CBA?

1:11:00Or do you say, well, I liked it then, but I wouldn't buy it now. Okay, well, I won't. That's the mental exercise to help you with that. That's the endowment effect. We think, you know, psychological, behavioral psychology, again, here I am again. The endowment effect is we value things we own more than if we didn't already own them. They've done great experiments on it, coffee mugs. You know, how much you would pay for the coffee mug is less than how much you would sell your coffee mug for, which is dumb. It's just really stupid. But when I say, again, I would do the same thing. Why? Because I'm human.

1:11:30The endowment effect is real. Human is the right word. Right, exactly. So I'm not bagging anyone else. I'm bagging myself and all of us. But know that is true so that you can help yourself deal with these sort of situations. Yep. Well said. Good luck to all shareholders out there. I hope it resolves in your favour. Oh, may the odds be forever in your favour, as they say on whatever that show was, Hunger Games. Yes. Mate, I reckon we've probably got through quite a bit today. A bit of international, a bit of local, a bit of investing psychology and a bit of a, be careful of star. Who knows? Again, speaking of 10 years' time, maybe Star is a thriving enterprise.

1:12:03They managed to get some funding. They dig themselves out of a hole and everything's okay. There are a range of possible future universes and some AI in there. Who knows what the future looks like? But I don't know. That's the most likely outcome. More presidential meme coins, which is something else we read out of time. We'll do that another week. We will. Don't buy meme coins. That's fun. All right. Don't buy meme coins. Don't buy meme coins. That's the short version. On that happy note, from a sitting president. Maybe especially. On that happy note, we will end this particular conversation.

1:12:35Will you join me on Sunday, mate? Yeah, looking forward to it, mate. I'm curious to see with no internet, how you've managed to entertain yourselves and keep yourself fit in the interim. So I will leave that hanging and we will talk on Sunday. Until then, fuller. Cheers. The Motley Fool and people appearing in this program may have positions in the companies mentioned. General advice only. Please speak to your financial professional to understand how it may pertain to your situation. Subscribe to the free newsletter at fool.com.au forward slash listener. The Motley Fool operates under Financial Services Licence 400691.