Opinion: Yes, artificial intelligence is fuelling a bubble, and it will eventually burst

John Rapley is an author and academic who divides his time among London, Johannesburg and Ottawa. His books include Why Empires Fall (Yale University Press, 2023) and Twilight of the Money Gods (Simon and Schuster, 2017).

The AI revolution is real and will transform the world. But that doesn’t change the fact that it’s probably still a bubble that will eventually burst.

In the meantime, it’s driving investors wild. Unlike shares in Canada, which have nudged ever so slightly upward over the past year, America’s S&P500 index has gone into orbit. The so-called Magnificent Seven, those big companies which are expected to profit most from artificial intelligence and have driven almost all of the U.S. market’s spectacular gains over the past year – Amazon AMZN-Q, Apple AAPL-Q, Alphabet GOOGL-Q, Meta META-Q, Tesla TSLA-Q, Microsoft MSFT-Q and Nvidia NVDA-Q – are together now worth more than the stock markets of every other country on the planet.

Leading the charge is Nvidia. A producer of software and designer of the chips powering the AI revolution, its share price had already reached stratospheric heights, more than tripling over the past year. When it released its quarterly earnings on Wednesday, it blew through the already sky-high expectations investors had for it. Needless to say, its share price kept rising, and it ended the week with a total market capitalization equivalent to Canada’s annual output. Yep, you read that right: one company is worth as much as a whole country.

That may say something about the Canadian economy, though that’s for another day. The question for now is, how long can this continue?

Those who say this time is different point to the many ways AI will change the way we live and work, ushering in a new industrial revolution of rising productivity growth, such that the valuations we’re seeing will justify themselves in time. Nvidia’s rapidly increasing sales do indeed seem to warrant the enthusiasm for the stock, at least for the time being.

But why that enthusiasm should transfer to the rest of the Magnificent Seven, let alone the economy as a whole, isn’t yet clear.

The dot-com bubble at the turn of the millennium produced the internet, smartphones and social media and created companies such as Amazon, Meta and Google which have profoundly altered society and the way we work, live and shop. However, it barely affected the productivity growth of the wider economy. Once it became apparent that the dot-com emperor had no clothes, that bubble burst.

Until we begin to see clear signs that AI will have the sort of broad-based productivity effects that, say, electrification had in the last century, we can probably assume it will follow a course similar to the dot-com bubble.

Moreover, this mania didn’t come out of nowhere. Over the past few years, the Federal Reserve pumped so much money into the financial system that it inflated a series of bubbles – crypto, meme stocks, housing. Market analysts have tended to focus on interest rates when discussing monetary conditions, and indeed complain frequently that central banks are taking too much money out of the economy. Yet what the Fed takes with one hand, it gives back with another. It’s still pumping money into the markets, using means like the liquidity measures it put in place during last year’s regional bank run. As a consequence, financial conditions remain as loose as they were two years ago.

So regardless what the long-term effects of AI will be, a bubble was ready to inflate. The problem for the Fed, though, is that this asset price inflation is now complicating its battle against consumer price inflation. Not only are shelter costs driving up other prices, but surging shares are swelling the retirement pots of older Americans, many of whom are cashing in and taking early retirement. The consequent reduction in labour supply is further tightening a market in which wages are rising faster than productivity.

Meanwhile, ballooning AI stocks are creating a risky amount of market concentration. In small markets, it’s not unusual for a handful of companies to dominate share trading – in Denmark, one company alone accounts for more than half the stock market’s capitalization. But in the United States, where 10 companies now account for more than a third of the stock market’s value, it’s rare. An awful lot is riding on a small number of companies making spectacular returns.

So it might not take a lot to prick this bubble. Many market analysts say this is a good time to buy because interest rates will come down this year, further supporting the rallies. In fact, though, underlying conditions suggest inflation has bottomed in the United States and upward pressures on prices may soon resume building.

Should it become apparent at some point that interest rates aren’t coming down, or should one of the Magnificent Seven fail to meet the outsized expectations investors have formed for it, the narrative could shift as suddenly as it did at the peak of the dot-com boom.

So buckle in, this ride is probably going to get even wilder.