Latest / America's Gold Authority Podcast / The AI Bubble: Boom, Bust, or Breakthrough?
Transcript
- 0:00SPEAKER_00: The views expressed in this video are those of the panelists and do not necessarily reflect the views of U.S.
- 0:05SPEAKER_00: Money Reserve.
- 0:11SPEAKER_02: Welcome to America's Gold Authority Podcast.
- 0:13SPEAKER_02: My name is Mike Barnes.
- 0:15SPEAKER_02: AI is the topic today.
- 0:16SPEAKER_02: There's been some selling off of AI stocks.
- 0:19SPEAKER_02: Is it a market correction or is the bubble bursting?
- 0:22SPEAKER_02: Let's go to our experts and find out.
- 0:24SPEAKER_02: Philip Deal is here, so is Brad Chastain.
- 0:26SPEAKER_02: And Philip, I want to start with you.
- 0:28SPEAKER_02: As the 35th director of the United States Mint and the president of U.S.
- 0:32SPEAKER_02: Money Reserve, are you a pessimist or an optimist about the future of AI?
- 0:37SPEAKER_01: I am an optimist, but for the interim, I'm a pessimist because I think we're going to go through a shakeout phase.
- 0:45SPEAKER_01: It's going to be very disruptive for the American economy.
- 0:49SPEAKER_01: I think there are going to be a lot of jobs lost for a number of reasons.
- 0:55SPEAKER_01: And I think there are going to be big changes in stock valuations.
- 0:58SPEAKER_01: There are going to be some winners and there are going to be a lot of losers in this.
- 1:04SPEAKER_01: And so I think it's really important for people to be ready for that transitional period, know what to look for as we approach that and be prepared for it.
- 1:15SPEAKER_02: Brad Jastain, Director of Education for U.S.
- 1:18SPEAKER_02: Money Reserve, same question for you.
- 1:20SPEAKER_02: Optimist or pessimist?
- 1:21SPEAKER_03: I think we're on the same page there.
- 1:24SPEAKER_03: That's certainly disruptive.
- 1:25SPEAKER_03: I don't think there's any doubt that there's tremendous value in economic value in what is being created, and it's going to be a revolutionary sort of evolution over time.
- 1:38SPEAKER_03: But there's a way to get there.
- 1:40SPEAKER_03: Folks are still learning how to really use it.
- 1:43SPEAKER_03: The technology is developing so fast.
- 1:45SPEAKER_03: The difficulty is knowing exactly where the revenue streams lie and where the productivity ultimately lies at the end, even though it's undeniable that it'll happen.
- 1:56SPEAKER_03: And right now, to Philip's point, the the investments have been grandiose.
- 2:02SPEAKER_03: I don't think is an understatement in terms of the levels.
- 2:05SPEAKER_03: And so without knowing what those revenue streams, that's that is a tremendous amount of risk.
- 2:11SPEAKER_03: And I think Philip, you mentioned something I don't remember the numbers or where it was about the fact that it could be economically impossible for revenues to ever be reflective of the amount of investment that's going in.
- 2:27SPEAKER_01: Yes, I I think that's right.
- 2:29SPEAKER_01: And that's one of my biggest concerns is uh JP Morgan has estimated that already 1.9 million,$1.9 trillion has been invested in AI.
- 2:42SPEAKER_01: And of course, more money is being poured into AI every day.
- 2:47SPEAKER_01: We we recently got numbers from Nvidia, and we see that that is still occurring.
- 2:54SPEAKER_01: The market is hot for uh AI stocks, and um and the prospect of being able to get that amount of money back out of this investment to cover that$1.9 trillion and earn a return on it is just looking very long shot.
- 3:16SPEAKER_01: And those estimates were looking at a possibility of only uh 0.8 uh trillion dollars in in a return versus maybe$1.9 billion.
- 3:28SPEAKER_01: So even under the best of circumstances, we'd only get that$1.9 trillion that's already been invested, get that money back.
- 3:39SPEAKER_01: And the route between today and the time that we would get that money back, where investors would get the money back, uh, is a long time coming, and there's a uh there's a lot of variables in work about whether or not we'd ever get it back.
- 3:58SPEAKER_03: It seems kind of crazy that you know the smartest people on earth, some of which are doing this, engaged in this behavior.
- 4:06SPEAKER_03: But it's it's uh it's kind of like uh a modern version of the old prisoners prisoner's dilemma.
- 4:12SPEAKER_03: If you think about the most basic example in uh game theory where a couple bank robbers are pitted against one another to decide, you know, uh if if one is going to confess and route the other and get a better deal, or it turns out in that dilemma, if people remember that, that their sort of best case overall is if they both remain silent, but because of the incentive structure, each individually is always best confessing and routing out the other one.
- 4:44SPEAKER_03: Because they either end up with the best possible scenario for themselves or avoid the worst possible scenario that the other person does it.
- 4:50SPEAKER_03: So in the end, they both end up engaging their own self-centered behavior and ending up with a worse deal than they would have been able to get if they uh they cooperated.
- 4:59SPEAKER_03: And it's very similar here because these big AI companies are banking their future on it.
- 5:05SPEAKER_03: And you could think they could coordinate and say, we're gonna go at a reasonable pace where we can make sure that our investment is going to be matched by expected revenues at some point in the future, but that's that agreement piece that is not really in any one company's best interest.
- 5:21SPEAKER_03: The rational decision for each one is to win that race because it is going to be so big in the end.
- 5:28SPEAKER_03: And if they cooperate, the other one cheats a little bit, they're out of the game.
- 5:32SPEAKER_03: They're irrelevant at some point.
- 5:34SPEAKER_03: But if they agree and they're the one that invests a little bit more, they win the race, the other one is out.
- 5:40SPEAKER_03: So it seems like a modern-day prisoner's dilemma there where they're acting rationally, despite the fact that the overall situation seems irrational.
- 5:49SPEAKER_01: We have, we really have four alliances of companies that are competing with each other.
- 5:57SPEAKER_01: And then there's some other players at the periphery.
- 6:00SPEAKER_01: Maybe they become uh more substantial.
- 6:03SPEAKER_01: Some of those players at the periphery are very big names with deep pockets, but they're not the core players right now in this market.
- 6:15SPEAKER_01: And those alliances around those companies are uh see in the future an enormous potential for profits.
- 6:25SPEAKER_01: It's just really, it's mind-boggling what the possibilities are.
- 6:30SPEAKER_01: They're also looking at a marketplace in which there seems to be unlimited capital that is flowing into them.
- 6:37SPEAKER_01: We see that right now, especially with Nvidia, but with some of the other companies too, OpenAI, and and now we have uh uh Elon Musk coming forward with a uh a big down payment from his own pockets and and from his investors.
- 6:57SPEAKER_01: And so the question is, is there room in the end, once the market settles out, is there room for all of these players in this market?
- 7:08SPEAKER_01: And the answer to that has got to be no.
- 7:11SPEAKER_01: There's not room for four players or so, maybe one or two.
- 7:16SPEAKER_01: And the and it's a this is a mad rush for economies of scale and scope.
- 7:24SPEAKER_01: And the companies that get to those economies of scale and scope will have a big advantage, and there's going to be intense competition, including price competition, getting to that point to be the first one at the finish line.
- 7:40SPEAKER_01: And part of the implications of that is price competition means zero or negative profits during that period of time.
- 7:50SPEAKER_01: And how are they going to pay back the$1.9 trillion?
- 7:56SPEAKER_01: You know, how are they going to generate a return on that?
- 8:00SPEAKER_01: And what has happened most recently, this is a new development, is that the capital that has been pouring in for the last several years into the IAA companies now is not sufficient to fund that race.
- 8:16SPEAKER_01: And so they're going into credit markets and they're issuing their bonds.
- 8:22SPEAKER_01: And they're going out and suggesting, they're actually going out and suggesting that maybe the federal government ought to stop in, step in, and backstop some of these investments.
- 8:35SPEAKER_01: It's madness how uh this feeding frenzy is progressing.
- 8:41SPEAKER_01: And the other thing we have to bear in mind, there's a human element to this too.
- 8:46SPEAKER_01: We all have watched the egos of these, almost all of the men, who have clashed and begun to compete with each other, whether it's in air, you know, in rocket travel or whatever.
- 9:05SPEAKER_01: They uh they have tremendous ego steep stakes.
- 9:10SPEAKER_01: And they have hundreds of billions of dollars or access to trillions of dollars that they can bring into this race.
- 9:20SPEAKER_01: Now, we saw something like this happen with dot-com in the late 90s that was entirely different, much smaller scale.
- 9:31SPEAKER_01: And the technology that those companies were putting into the ground, mostly uh fiber optic cables, even when the bubble burst, that technology was still in the ground and still completely usable.
- 9:45SPEAKER_01: It didn't go to waste.
- 9:47SPEAKER_01: But that's not the case today.
- 9:50SPEAKER_01: The enormous amount of money that's going into the hardware and modeling systems uh of AI are uh have a life cycle of about two to four years, which means once, you know, once we get down the road another two to four years, all of that's gonna have to be replaced.
- 10:12SPEAKER_01: And uh so, you know, tremendous, we're looking at tremendous second round of capex capital expenditures to fund the industry.
- 10:23SPEAKER_01: And then that opens the question of how fast is this revenue, um, this revenue track going to run?
- 10:32SPEAKER_01: Is it ever going to be able to keep up with cap CapEx?
- 10:36SPEAKER_02: Well, Philip, I want to go back to the the capital, that$1.9 trillion with the T dollars.
- 10:42SPEAKER_02: That's more, I think, than the entire GDP of the state of Florida.
- 10:45SPEAKER_02: What in the world are they doing with that money?
- 10:49SPEAKER_01: It's an incredible number.
- 10:51SPEAKER_01: And actually, the forecast that within the next couple of years, that's going to become$3 trillion, which I think is getting close to the GDP of New York.
- 11:03SPEAKER_01: Um, so they are spending this money on, first of all, the data centers, the physical structure of the data centers, which are huge, the air conditioning systems that are necessary to keep them cold enough for the technology to continue working at its peak efficiency.
- 11:22SPEAKER_01: They're talking about the Nvidia chips that provide the compute power for these systems.
- 11:30SPEAKER_01: They're talking about the cost of these really sophisticated compute models and um and other kinds of software technology that is constantly increasing in its sophistication and its compute power.
- 11:49SPEAKER_01: And um we know this in our own experience with technology, with the iPhone, how quickly we move from one generation of capabilities to the next generation of capabilities.
- 12:01SPEAKER_01: And before long, your phones or your camera on your phone is obsolete.
- 12:07SPEAKER_01: What we're talking about here is a trillion-dollar iPhone that is, you know, that is uh over uh turning over every six months.
- 12:17SPEAKER_01: So this is a much faster kind of uh next gen uh driver for this kind of technology.
- 12:26SPEAKER_01: So um, and so you know, the the data centers themselves, they have a 10, 15, 20 year lifespan.
- 12:35SPEAKER_01: They're they're building.
- 12:36SPEAKER_01: And the air conditioning systems, they have maybe a five to 10 year lifespan.
- 12:42SPEAKER_01: But the the the NVIDIA chips, they will burn out in a matter of a year, year and a half.
- 12:49SPEAKER_01: And um and then they'll become obsolescent obsolescent anyway, because uh the new chips that come out will be so much more powerful.
- 13:01SPEAKER_02: Brad, Philip mentions Nvidia.
- 13:03SPEAKER_02: Uh, some numbers came out about Nvidia this past week.
- 13:06SPEAKER_02: Does it give even more optimism about AI because of those numbers?
- 13:11SPEAKER_03: Well, it it still is the same thing.
- 13:13SPEAKER_03: I mean, we know that the investments have been going in, and so their numbers, to answer your question, they beat estimates, showing there's still a tremendous amount of investment.
- 13:22SPEAKER_03: And they're the recipient of those investments, this Philip just described, right?
- 13:26SPEAKER_03: So um it it proves what's happened in the past.
- 13:30SPEAKER_03: It does not necessarily be it's not indicative of what we see in the future, if that makes sense.
- 13:36SPEAKER_03: Are there a lot of AI jobs out there?
- 13:38SPEAKER_03: There are a lot of AI jobs.
- 13:41SPEAKER_03: Um, and more and more, what in it's not so much in AI.
- 13:45SPEAKER_03: There are a lot of jobs being created in the AI space, but what we're also seeing is a transition for jobs outside of any AI sector in normal companies, people with AI skills, to be able to leverage those technologies for more uh productivity.
- 14:02SPEAKER_03: So it is shifting the workforce a bit outside of the AI sector, but it's also shifting it from away from things that AI is, of course, replacing.
- 14:13SPEAKER_03: And it's slow at this point.
- 14:15SPEAKER_03: There's nothing dramatic at this point, but we are seeing an impact directly from AI in entry-level um type jobs, entry-level programming, um, computer programming, entry-level clerical kinds of things, uh, customer service types of jobs, and in those areas.
- 14:34SPEAKER_03: Um, but then the people that are being hired in those areas are those that are more familiar with using it.
- 14:39SPEAKER_03: So it's shifting how the workplace works.
- 14:42SPEAKER_03: But there's been some studies, uh, Bank of America, Yale did one as well, that show that so far at least the disruption is pretty much on par with other technological disruptions in history.
- 14:56SPEAKER_03: But that at this pace this is developing and the and as uh as fast as AI is becoming more capable, um, that is uh increasing uh pretty quick.
- 15:08SPEAKER_03: So it's hard to imagine that it'll stay in that place.
- 15:12SPEAKER_03: The the ones that are seem to be the most protected at this stage are the really the the uh trade type jobs.
- 15:20SPEAKER_03: You know, a good friend of US Money Reserve, Mike Rowe of the well-known show Dirty Jobs, uh, has a foundation about the trades and supporting people getting into the trades.
- 15:30SPEAKER_03: And those are the things that right now at least are the most insulated from that from that impact.
- 15:36SPEAKER_03: It's really more of the white-collar entry-level jobs that we're seeing it work its way through.
- 15:42SPEAKER_02: I'm seeing some new lingo out there, Philip.
- 15:45SPEAKER_02: AI companies make sense, but AI adopters, AI alliances, what does all this mean?
- 15:51SPEAKER_02: And what do I need to know about it?
- 15:53SPEAKER_01: Okay, AI adopters are the companies that uh use the products and services that are being produced by the AI companies.
- 16:04SPEAKER_01: So you have companies that are outside of the AI sector.
- 16:08SPEAKER_01: They may be in the tech sector or not in the tech sector, but they're going out and buying these products and services, AI products and services that allow them to become much more productive.
- 16:25SPEAKER_01: Uh it's like uh the laptop computer or desktop computer was 30 years ago, how much more productive office work became when they came into the marketplace.
- 16:38SPEAKER_01: Now, the way that these companies are increasing their productivity is by being able to do new things they haven't been able to do before, therefore making more profits, and by cutting headcount, cutting jobs.
- 16:56SPEAKER_01: And ultimately, that's one of the key ways you increase productivity in any company, is to be able to do more with fewer people.
- 17:07SPEAKER_01: Now, um, so that's really the crux of the matter here, because what it looks like right now is that the AI companies, in order to generate a return on this$1.9 trillion, they're going to have to be selling to people who will cut lots of jobs.
- 17:32SPEAKER_01: That's the only way they're going to be able to turn a profit.
- 17:35SPEAKER_01: Right now, today, MIT issued a study within the last month or so that said all of the investment that has already gone into those AI adopters, all of the investment they have made is making almost no impact whatsoever on their productivity or their or their profits.
- 17:56SPEAKER_01: So there's open questions about how much, and we're in the early stage.
- 18:01SPEAKER_01: There's no question about that, we're in the early stage, but those companies are going to have to ramp up, considering how expensive this technology is, they're going to have to ramp up their productivity.
- 18:13SPEAKER_01: And I think that's going to mean a major slashing of employment among those AI adopters.
- 18:20SPEAKER_01: Now, the AI alliances, this is this is something that's you really need to have a roadmap to have a visual of this in front of you to understand it.
- 18:31SPEAKER_01: But you have these four centerpieces of AI companies, and then you have this constellation of other companies around each of them.
- 18:43SPEAKER_01: And some of these other companies, they're providing uh products and services uh and funding uh and other contractual arrangements that support those four companies.
- 18:58SPEAKER_01: And there are companies that are within an alliance that are in multiple alliances.
- 19:05SPEAKER_01: So what you really need to think about is that those companies are that are involved with multiple uh you know centerpieces, multiple main AI companies are hedging their bets.
- 19:18SPEAKER_01: They don't know who's gonna be the winner.
- 19:20SPEAKER_01: So they're providing services to multiple companies or drawing funding from multiple companies.
- 19:28SPEAKER_01: And over a period of time, probably not very long, we're gonna begin to see those companies uh go through a shakeout where some of them become very clear winners and others will become what appear to be losers, and the losers are gonna find it more and more difficult to raise the capital to stay in the race.
- 19:53SPEAKER_02: So, guys, if the S P 500 is vulnerable, is gold something that someone should look at to diversify?
- 20:01SPEAKER_01: Yeah, absolutely.
- 20:02SPEAKER_01: I in fact, I think gold is the primary way to diversify and to hedge this market.
- 20:08SPEAKER_01: Um, if we're talking about, and we are talking about a situation in which the SP 500 has become so dependent upon the performance of not just the technology set uh sector, but seven AI companies within the technology sector.
- 20:28SPEAKER_01: So what we're talking about is seven companies uh representing 38% of the market value of the S P 500, and the other 67% uh is or 62% is uh is uh attributed to the other 493 companies.
- 20:51SPEAKER_01: So we have this heavy dependency.
- 20:54SPEAKER_01: And if those Mag 7 companies bring down all of the S P 500 because of index funds, uh that is a very likely scenario, then you want to have an asset that has a track record of performing well in tough economies.
- 21:14SPEAKER_01: And gold really plays that role, has played that role, and it's outside of the uh economic system, the financial system.
- 21:25SPEAKER_01: And as other dominoes fall, like we saw in the 2008 financial crisis, certainly we saw in the 2000 in the 1929 crash, uh you you want to have your money outside of that financial system.
- 21:41SPEAKER_01: And we're seeing it, and we've been seeing it for the last two to three years, is that as these risks mount and spread across uh the political world and in the financial and economic world, uh, gold has rallied as a way of protecting wealth.
- 22:00SPEAKER_02: And Brad, you mentioned landmines.
- 22:02SPEAKER_02: It's good to have a safe haven like gold to walk through those landmines.
- 22:07SPEAKER_03: Yeah, absolutely.
- 22:08SPEAKER_03: I mean, and that that's what we talk about all the time is being diversified to be able to handle whatever lies ahead.
- 22:15SPEAKER_03: Because we're talking about risks that are out there, but you'll notice Bill and I never say these are absolutely things that are going to happen.
- 22:24SPEAKER_03: It's in investors out there, uh financial folks who are always convinced that there is one possible outcome for a scenario end up doing rather poorly, right?
- 22:36SPEAKER_03: There it's always a probability distribution of what can happen.
- 22:41SPEAKER_03: And so you want to know, think through how do you is it, do you think 70, 80%, 50% is likely to happen?
- 22:47SPEAKER_03: Even if it's 50-50 that the a bubble pops in the next year, two, five years, what are the downside losses versus potential gains, right?
- 22:56SPEAKER_03: Those are all things that we have to think about.
- 22:57SPEAKER_03: Because if it's a, you know, the market has a correction like 2008 and it's down 50%.
- 23:03SPEAKER_03: But because markets are so elevated, even if it doesn't pop, it just doesn't go up very much.
- 23:08SPEAKER_03: It's not a lot of upside for downside with a 50-50 shot in that scenario.
- 23:12SPEAKER_03: That's kind of how we have to think about things overall.
- 23:14SPEAKER_03: What are the probabilities?
- 23:16SPEAKER_03: And having different types of assets that will be able to weather different scenarios of outcomes is just it's so important.
- 23:24SPEAKER_03: You can't have all of your eggs in one basket.
- 23:26SPEAKER_03: And so having something like gold that responds very differently than other assets provides that diversification, uh, that protection.
- 23:34SPEAKER_03: And that's why we've shown and talked about time and time again how much it has helped improve the overall performance of a portfolio over time from both on a risk-adjusted basis, meaning what returns have you gotten versus the amount of risk that you're taking as a whole.
- 23:50SPEAKER_03: And um, and that's something if anybody is interested in, we have it in our materials.
- 23:55SPEAKER_03: You can compare different portfolios and how they've performed over time with and without gold, and you can see that it's drastically reduced the risk in a variety of economic conditions while actually helping returns improve.
- 24:09SPEAKER_03: Um, but even if it was the same returns overall with less risk, that's still a compelling reason to have an allocation.
- 24:18SPEAKER_01: Yeah, and I think I think following up on Brad's point, when uh clients go in and talk to their financial advisors, I hear this all the time, uh they are they also often in their they want to talk about gold.
- 24:31SPEAKER_01: And a financial advisor will often push back and say, well, but do you want to give up?
- 24:38SPEAKER_01: You know, do you want to sell some of your stock and give up the potential return on those stocks and uh and put it in a uh in a product, a financial asset that doesn't generate a uh a return.
- 24:55SPEAKER_01: And the thing to bear in mind and the pushback on gold is as a matter of fact, gold has outperformed the SP 500 over the last 25 years, has over the last 10 years.
- 25:11SPEAKER_01: You can look at the uh the peak stock performers of the Mag 7.
- 25:18SPEAKER_01: I mean, that's what everybody talks about, how the Mag 7 is driving the SP 500, uh, the gains in the Mag 7, and gold has outperformed the Mag 7 over the last two years.
- 25:32SPEAKER_01: And last year in particular was a huge year.
- 25:35SPEAKER_01: Today, this year has been a slower year for the Mag 7.
- 25:39SPEAKER_01: But gold has outperformed it.
- 25:41SPEAKER_01: So it is not the case that, like it used to be, it's not the case anymore, that you give up potential gains uh by selling stocks and buying gold in order to diversify your account.
- 25:59SPEAKER_02: Well, that's where you love something where you can put your money somewhere where there's very little risk, but a lot of reward.
- 26:04SPEAKER_02: Is that right, Brad?
- 26:06SPEAKER_03: Yeah, absolutely.
- 26:08SPEAKER_03: And you can run back, anybody watching can go back to previous episodes, and we've talked about fundamental reasons why that is, with the debt growth globally, the amount of money that has to continue to increase relative to that.
- 26:21SPEAKER_03: And uh, and if we do have an economic slowdown, guess what?
- 26:24SPEAKER_03: That's absolutely that's more debt that governments take on to try to stimulate the economy, which requires more so-called money printing to keep the ball rolling.
- 26:34SPEAKER_03: And that's what gold over time, not tick for tick, but over time responds to.
- 26:40SPEAKER_03: And go, and that has grown at a faster rate than inflation.
- 26:44SPEAKER_03: And as a consequence, since then, gold has outperformed even the normal inflation rate and stocks and all of those other things.
- 26:51SPEAKER_03: So absolutely.
- 26:53SPEAKER_02: Well said.
- 26:54SPEAKER_02: Brad Chastain and Philip Deal, thank you very much.
- 26:57SPEAKER_02: Thank you, Mike.
- 26:58SPEAKER_02: Thank you, Mike.
- 26:59SPEAKER_02: Well, that's all the time we have for today's episode.
- 27:02SPEAKER_02: On behalf of all of us here in the U.S.
- 27:03SPEAKER_02: Money Reserve Studios, thank you for tuning in.
- 27:06SPEAKER_02: If you have any questions or you want to learn more about today's episode, you can call the number in our episode description and speak with one of our friendly account executives.
- 27:16SPEAKER_02: They will be more than happy to answer any questions you may have about precious metals and the strategic role they can play in achieving a secure and comfortable financial future.
- 27:26SPEAKER_02: I'm Mike Barnes.
- 27:27SPEAKER_02: We'll see you next time on America's Gold Authority Podcast.
- 27:36SPEAKER_00: The markets for coins are unregulated.
- 27:38SPEAKER_00: Prices can rise or fall and carry some risks.
- 27:40SPEAKER_00: The company is not affiliated with the U.S.
- 27:42SPEAKER_00: government or the U.S.
- 27:43SPEAKER_00: Mint.
- 27:44SPEAKER_00: Past performance of the coin or the market cannot predict future performance.
- 27:46SPEAKER_00: Offer void where prohibited.