Photo: GORODENKOFF PRODUCTIONS/SCIENCE
More and more, finance experts are predicting that the AI bubble is getting ready to pop - so what does that mean for Kiwi investors, and for our economy?
It took the United States 40 years to build its interstate roading network. The same amount of money spent on that network - inflation adjusted - has been spent in just three years on building Artificial Intelligence data centres.
Fisher Funds' Harry Smith uses the comparison to illustrate the size and pace of growth of AI and its future demands for infrastructure.
He points to five of the famous tech companies - Meta, Google, Microsoft, Amazon and Oracle - which together plan to spend US$3 trillion over the next five years on AI infrastructure, specifically the data centres which store the information, called the cloud.
The trouble is, says Smith, their combined revenue is US$1 trillion, excluding Amazon's retail side, and many are beginning to ask how they will fund the massive infrastructure builds.
At the same time, their share prices have been soaring as investors bet on the potential of the companies in the future, leading to record valuations in the likes of chipmaker Nvidia. It became the first company in the world to break through US$5 trillion valuation but has see-sawed in a volatile market.
"The amount of money that the market expects to be generated over the next five years is quite huge and that's where a lot of the bubble chat is really coming from," says Smith.
The bubble chat he's talking about came to a head a week ago when Michael Burry, who inspired the book and movie The Big Short announced he was short selling on chip maker Nvidia and data analytics company Palantir. By short selling, he was taking a bet that their share price will drop.
Burry has been called a "tireless perma-bear", a contrarian investor who predicted the 2008 financial crisis.
Not all bubbles are the same
He isn't alone with his warnings about market mania. Meta's Mark Zuckerberg, Amazon's Jeff Bezos and Open AI's Sam Altman have also cautioned about the overblown stock prices alongside the Federal Reserve, International Monetary Fund and the Bank of England.
Smith, who runs a global equity portfolio at Fisher Funds, says it is important to separate the hype from the technology when trying to understand what is happening to AI shares.
The technology is very real, he says and is already in use in the medical profession with diagnoses and medicines.
"Further on, we're seeing developments around AI being used in drug development ... the creation of new drugs and trials.
"When I'm talking about hype, people are saying 'hey the earnings in these companies are going to be massive in the future, so we're going to buy the share price' and that's when the bubble comes into question because [other] people are saying 'no, the expectation of the earnings growth that these companies are going to deliver is too great and therefore we are in a bubble in this cycle.
He uses Open AI as an example of a company with big investment plans and comparatively small revenues.
"They currently have about $US13b in revenue this year, they currently have commitments of US$1.5 trillion of investment that is going to be needed to be built. So on the US$13b, albeit it is growing really quickly, but they are losing money at one of the fastest paces ever in Silicon Valley," says Smith.
But Smith says not all bubbles are the same. Banking bubbles are bad and cause society a lot of pain, as seen in the 2008 global financial crisis. Industrial or technology bubbles can benefit society when the dust settles.
"This whole question around AI is what is the return on investment because we know the investment is huge, we know that the revenue or profits that are currently generated are very small and so it is about trying to find a path to that return.
"How quickly can revenue and profits grow from these AI businesses to support the investment that is currently being made? That is the crux of the question around whether or not there's a bubble."
RNZ business editor Gyles Beckford says the way investors have been behaving with AI shares is the same as 25 years ago with the dot-com boom and bust.
"There's the same irrationality that attends all these things. You have the studious people who do their homework, who believe that they've crafted the right strategy, that they're in the right stocks and they will go with it and will be there for the long term.
"They're the survivors, I shudder to use the word cockroaches but they will come through the explosions. There are those who suffer from FOMO, fear of missing out."
They are the ones making spur of the moment, irrational decisions, he says.
"History tells us that when you have surges like this in particular assets, particular stocks or commodities, there is a cleanout."
Check out how to listen to and follow The Detail here.
You can also stay up-to-date by liking us on Facebook
