Expert view: could AI be heading for a dotcom moment?

There are growing concerns that the AI boom could eventually turn into a bubble, particularly after the extraordinary gains seen across the biggest technology companies. The parallels with the dotcom era – when investors piled into firms with big digital dreams and valuations raced far ahead of reality – appear to be harder to ignore.

Nearly $33 trillion in market value has been added to the US stock market since the AI boom began, with the vast majority coming from companies with futures tethered to the technology.

However, with valuations elevated, enormous infrastructure spending and growing concerns about the safe deployment of increasingly powerful systems, the pressure points in the AI story are increasing, amid a growing debate among America’s top AI labs over whether AI development is moving too quickly.

In my view, the impact of any potential regulatory response is certainly not priced in and – should such a response materialise – it may cause significant volatility.

That said, AI tools should allow for a substantial leap in productivity for firms that opt to use them to their fullest.

So, while it is difficult to foresee a big slowdown in AI demand, governments may look to rein in how quickly the technology and its infrastructure can expand, and that may mean investors reassess their growth expectations.

A correction could be painful, but it could also be a useful sorting hat, separating the companies with genuine staying power from those simply riding the hype.

Not all tech companies are created equally. Just like the dotcom boom and bust, it is conceivable there will be casualties and survivors. Some companies will probably fail to live up to the AI hype. Others – with opportunity stamped into their DNA – may be fit enough to weather a sudden drop in confidence. The current group of AI market leaders appear significantly stronger operationally and financially than the bubble stocks of the dotcom era.

Should the AI bubble burst, the FTSE 100 may not take as big a hit as the US market. The UK index has far less exposure to the giant technology companies that have driven much of the AI-fuelled rally, and is more heavily weighted towards banks, energy companies, miners, healthcare and consumer staples. That could provide some shelter if investors suddenly decide the AI emperor has rather fewer clothes than they thought.

But the FTSE would certainly not be immune. A sharp sell-off in US technology shares could trigger a much broader flight from risk, pulling down markets around the world, while UK companies benefiting from the AI spending boom or supplying the infrastructure behind it could also feel the chill.

There could be implications for the bond market too, and for investors whose portfolios include corporate and government debt. The extraordinary build-out of data centres and computing infrastructure is no longer being funded simply from the cash piles of the biggest technology companies. Increasingly, debt is being used to finance the AI arms race, with the major hyperscalers issuing huge amounts of corporate bonds to fund their expansion.

If enthusiasm for AI starts to fade, investors could reassess not only the value of the companies issuing the debt, but whether the infrastructure being financed will generate enough revenue to justify the borrowing. And this could cause fresh volatility in the bond markets too, particularly if investors start demanding higher returns to compensate for the risks. This could have implications for portfolio valuations which have been skewed towards bond investments.

There is a potential cushion: some of the infrastructure being built to meet AI’s huge electricity demands – from renewable generation and storage to grid upgrades – will likely also be needed for the wider electrification of economies and the net zero transition. So, while not all AI infrastructure spending would have the same value if AI demand disappoints, some of the investment could have a useful second life supporting the shift towards cleaner and more electrified energy systems.

Nobody can predict whether AI enthusiasm will keep building or eventually give way to a correction. What investors can control is how exposed they are to any single theme. A well-diversified portfolio should not be overly reliant on one technology, sector or market to deliver returns.

That means resisting the temptation to chase the biggest winners, however compelling the story may appear. Spreading investments across different asset classes, sectors and regions can help reduce that risk.

Investors should also keep a long-term perspective. Periods of volatility are an unavoidable part of investing, and markets have repeatedly recovered from sharp setbacks. The challenge is that recoveries do not always happen as quickly as investors expect, making resilience and diversification just as important as return potential.