AI: Bubble or Revolution?

I posit that AI is currently a bubble, but before I explain why I want to disclose my personal biases.

I am biased towards humanity, and in particular the poor and middle class, and my conjecture is based on the idea that the best way to measure the health of humanity is via economic prosperity.

History has shown us time and time again, without fail, that when the economy is booming, everyone gets richer, and in particular the poor and middle class.  

However, when the economy is stagnating or declining, only the rich get richer, and everyone else gets poorer. One of the permanent features of humanity we must learn to accept is that the rich will always get richer. The poor and middle class will only prosper during times of genuine economic advancement.

It is therefore my firm belief that economic advancement must be humanity’s number one priority. If there was a singular measure we could use to assess humanity’s progress it would be the number of children who no longer need to live under the tyranny of evil suffering and poverty.

With that in mind, lets explore why I believe AI is a bubble.

What is a bubble? Bubbles are characterised by asset prices that achieve extraordinary amounts of growth. And I’m not talking about above-average performance. I’m talking about performance that is several orders of magnitude higher than the long-term exponential averages. And what happens to asset bubbles? The same thing that happens to all bubbles. When they get too big, they burst.

In every financial market cycle in history, bar none, every time an asset class has defied a long-term exponential trend for long enough, it has corrected in spectacular fashion. And there is one key ingredient to a bubble, which usually occurs right at the time the bubble is about to burst and that stage that is called “mania”. Mania is when the sentiment of both institutional fund managers and weekend warriors alike, is such that asset prices will continue growing as they currently are. The manic investor assumes that the current state of affairs will either continue, or improve indefinitely.

Now you can show these manic investors just how bad macroeconomic fundamentals are. You can show them that economic growth is failing to outpace inflation, meaning that humanity is appearing richer on paper but poorer in reality. You could point out that almost every unemployment index in the world is heading in the wrong direction. You could show manic investors that average household savings are the lowest they’ve ever been. You could show evidence that the average household is barely treading water in a growing tsunami of consumer finance, credit cards and predatory lending such as afterpay and payday advances. You could show these people that the vast majority of the money that has landed in the stock market in recent years causing asset prices to swell was printed by banks and not earned through economic prosperity, resulting the highest leverages of leverage in history. You can show the manic investor that almost all stock market indices are priced way above long-term exponential trendlines, and that corporate earnings are almost triple their long-term averages. You could demonstrate it would only take a slight reduction or a downgraded forecast in corporate earnings, to send all major global share markets into a meltdown.

But those in the stage of mania care not for reality, and will always revert to their favourite trick: Chart analysis. When there are absolutely no positive macroeconomic indicators in sight, manic investors get extremely creative. I personally love hearing chart analysis, not because I find it valuable, but purely for entertainment quality. I liken chart analysis to the bloke at the pub with a schooner betting on the ponies on a Saturday morning with his form guide in hand. I find it adorable, because, like the guy at the pub, manic investors are able to draw some truly amazing conclusions from carefully selected segments of historical data. They’ll zoom in on section of a chart, and they’ll say things like, every time this shape happened in the past, the stock went on grow by x%, and given we’re seeing this shape again, it is basically guaranteed to continue, right?  

And that’s the thing with horse racing, crypto gambling and stock market trading. Historical trends just keep repeating themselves… until of course they don’t. And when the cycle changes, you always hear the same word to describe what happened and that word is “unprecedented”. Well I’m hear to tell you that that the current bubble is in fact NOT unprecedented. Current market cycles are very much precedented. Mania and ignorance have always been there. Always have been and always will be.

And this where the current mania cycle has gone next level.

One of my favourite mantras of manic investors right now is because artificial intelligence is unemotionally and efficiently driving most of the investment decisions on behalf of the world’s largest fund managers, that we won’t have a stock market crash. Really, are you telling me that artificial intelligence with its godlike knowledge of financial markets only knows how to bet long on the market but not short? (Betting short means making money when the price drops and short-selling is extremely profitable. Let’s hope AI doesn’t learn how to short-sell!). One of the craziest things I heard recently is tech company earnings will continue to grow exponentially because quote-unquote “everyone is spending money on AI and all of humanity will benefit from AI.”

Now this, my friends, is where the delusion sets in, and what I’m about to tell you is going to have that same “how the fuck did we miss this?” feeling that global fund managers felt when the dotcom bubble burst in 2001, and when subprime mortgages went bad in 2008.

Remember what I said at the start. During times of economic prosperity, everyone gets richer. During times of economic decline, only the rich get richer and everyone else gets poorer.

Hard truth time: Artificial intelligence only benefits the rich (at least for now anyway).

Think about this for a moment. Let’s ignore macroeconomic fundamentals and chart analysis. Let’s focus on the real world the way it currently is. Where are we spending our money right now and who really benefits from AI?

One of the things that the poorest 99% of the world are really struggling with right now is the cost of living. Let’s ignore the causes of the cost of living crisis right now – I could go on about that all day. Let’s also not speculate on what may happen in the future and let’s just focus on today’s reality.

By far, the biggest contributor to the cost of living crisis, and the average person’s biggest expense right now, whether you’re in Australia, America, Denmark or India, is housing. Whether you’re renting, owning, building, or renovating, you are paying more as a fraction of your disposable income than you’ve ever paid before. In fact, depending on where you live, the poorest 99% of the global population are paying 40 to 90% of their after-tax earnings just to keep a roof over their head.

What’s the next biggest expense for the average person? Food. I don’t think anyone would debate that their grocery bills right now are so high that it almost feels like a bad dream when they go to the supermarket or out for dinner with friends.

And the third biggest expense is of course, utilities. I’m talking about water, electricity and gas. It is simply astounding to compare today’s utility bills to what they were just 5 years ago.

The poorest 99% of people are spending the vast majority of their after-tax income on housing, food and utilities, and nowadays it is a luxury to have money left over afterwards. Of course, none of this should shock you, and all of this is supported by these pesky macroeconomic data points the manic investors are evading.

So, what do housing, food and utilities all have in common?

They are all primary production industries. You cannot have housing without land, and you cannot build without materials. Small tribes and villages may survive without infrastructure and agriculture, but cities and nations couldn’t. And you simply cannot have water or power without natural resources. Is it any wonder that these industries tend to perform very well during recessions while non primary production industries like banking and technology tend to get slaughtered?

And herein lies the questions I have for the all the manic investors out there who claim we are currently in an AI revolution:

How is AI currently making land more affordable?

Do we have robots extracting raw materials from the earth and distributing them around the globe, or is it still mostly humans doing that?

What about manufacturing and construction? I haven’t seen any robots swinging hammers on construction sites yet. Have you?  

Don’t get me wrong. It is all but certain that in the not-too-distant future, artificial intelligence will be directly inserted into primary production roles. But are we there right now? Absolutely not. Humanity will need time to adjust to this, and it won’t just happen overnight.  

In the meantime, I invite you to consider an alternative theory: AI is not good for the economy.

Not initially anyway. In fact, it will be incredibly destructive to the broader economy, benefiting only the rich and decimating much of the poor and middle class.

Granted AI has some amazing features right now. But what are these features and who are these features benefiting?

Don’t get me wrong, large language models such as ChatGPT are really impressive, but who benefits from it? LLM technology assists in research, consolidating a lot of complex information very quickly and is extremely effective in sales and marketing. But who benefits from this? Business owners and executives do, as they will be able to increase their bottom lines whilst significantly reducing human resources. LLMs are simply better researchers than humans. Graphics algorithms like Midjourney are of great assistance to anyone involved in art, design, entertainment and architecture, but again, all this does is increase profits for the people at the top while reducing opportunities for good old fashioned human creators. As for automations and workflows, there are countless algorithms and bots out there are currently increasing administrative and logistical efficiencies for a wide range of industries, but again, all this does is benefit the people at the top while reducing the demand for human resources.

In summary, artificial intelligence is very good at moving information. But is it doing anything to meaningfully increase primary production? What is AI doing to make housing more affordable? Can you eat or drink artificial intelligence? What is artificial intelligence doing to reduce the cost of powering our homes and infrastructure? What is AI doing to help the poorest people escape poverty? Nothing. Things are only getting more expensive, benefiting only the rich, while the poor and middle class get poorer.

With the diabolical state of our economy can we honestly say that we are experiencing an AI revolution? Until you can show me evidence that AI is causing real economic advancement, the kind of advancement that benefits everyone, I maintain that all the hype is characteristic of a classic bubble.

And as a final thought, one of the things I’m getting tired of hearing from AI enthusiasts is this: “AI will not replace your job, a human using AI will.” Whoever thinks this way clearly does not understand business. By far, the largest expense incurred by businesses, from small local businesses to NASDAQ giants is human resources. If a business no longer needs humans to achieve its desired level of productivity, it will stop hiring. It may even start firing, like most of the so-called “magnificent seven” of the NASDAQ are currently doing, with their biggest layoffs in history. If you are humanist like I am, you may have all the best intentions to continue hiring humans, but at what rate? As AI advances, hiring rates across most industries will slow down. And not because we don’t want more staff, it is because making an existing employee redundant is vastly different to never needing to hire them in the first place.

This is happening right now, with devastating effects to the economy.

So again I ask, how exactly, is the current state of affairs a revolution?

Our portfolios have been positioned in anticipation of the AI bubble bursting. We saw a small taste of what may come in the first week of August 2024. What was catastrophic for global financial markets barely registered as a blip in our “Tuna Salad Wrap”. We’ll continue our important work in this strategy.