Will the AI bubble burst?

Artificial intelligence (AI) megacap stocks have soaked up much of the stock market’s gains over recent years, but there are fears that this growth is increasingly resembling a bubble that’s fit to burst.

We’ve become used to the AI megacaps being the top stocks for investors over recent years, but top bankers from the likes of Morgan Stanley, JP Morgan and Goldman Sachs have voiced their concerns that AI stocks’ valuations are reaching bubble-like levels.

Legendary contrarian investor Michael Burry – famously portrayed by Christian Bale in The Big Short – allocated over 80% of his fund to short positions on Nvidia (NASDAQ:NVDA) and Palantir (NASDAQ:PLTR), according to its latest regulatory filing.

MoneyWeek

Subscribe to MoneyWeek today and get your first six magazine issues absolutely FREE

Get 6 issues free

Sign up to Money Morning

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Sign up

The week to 7 November saw the worst week since April for AI stocks, with the Nasdaq 100 index (predominantly composed of US tech stocks) falling by over 3% over the course of the week.

While the Nasdaq has since staged a partial recovery, it has not been smooth sailing. Shares in CoreWeave (NASDAQ:CRWV), an AI infrastructure company with close links to Nvidia, fell 16% on 11 November, following full-year revenue guidance that came in below analysts’ expectations.

According to CoreWeave the forecast downgrade is due to just one third-party data centre (out of 41 in its portfolio) being behind schedule, but it underscores how stretched valuations have nothing but good news priced in and that the slightest negative can lead to dramatic falls.

“The negative reaction… suggests some growing nervousness about valuations in the AI space,” said Russ Mould, investment director at AJ Bell.

Nvidia’s share price fell 3% on the same day as news emerged that Japanese investment giant SoftBank has sold all of its shares in the chipmaker.

“Corrections are a healthy and necessary fact of life in financial markets but investors will be wary of any signs this is turning into a pronounced sell-off,” said Mould. “For now, we are not really in that territory but the stakes are ratcheting up ahead of Nvidia’s third-quarter earnings update on 19 November.”

Is AI a bubble?

The tricky thing about stock market bubbles is that they can only be definitively identified in retrospect.

“We’ve heard quite a few loud warnings of a bubble recently, but I think the jury is still out,” said Craig Baker, chair of the Alliance Witan investment committee.

Baker highlighted a need to differentiate between “AI-exposed companies with strong fundamentals”, particularly the highly profitable big tech giants of the Magnificent Seven, and the flurry of smaller businesses riding in the wake of investor enthusiasm for AI.

“There may well be a bubble in the share prices of the latter, some of which don’t even have revenues let alone profits. And the valuations of some of the former are also starting to look a bit stretched,” said Baker. “But even if there’s enough evidence to worry about being in a bubble, there’s not enough to suggest it’s about to burst.”

The huge multiples at which AI stocks trade only makes sense in a world in which they continue to strongly outperform the market.

But that isn’t guaranteed to be the case. Nvidia’s sales growth slowed from 265% in Q4 of its FY 2024 to 53.5% just a year and a half later.

With its current share price effectively factoring in continued revenue growth for years, there are concerns that Nvidia might be overvalued.

“It’s possible the Magnificent Seven, or some of them at least, may continue outgrowing and outearning the rest of the global stock market for many years to come,” said Baker, “but even Sam Altman, boss of OpenAI and one of the leading evangelists for the new technology, thinks we might be in a phase where investors are overexcited about it.”

Risks to the AI model include competition for energy and computing resources, as well as “scarce AI talent”, with “regulatory and geopolitical uncertainty” overlaying this, said Ben James, investment specialist at Baillie Gifford US Growth Trust.

But these risks are being weighed against the long-term potential of AI. “The dotcom boom was based upon the premise that the internet would transform business, which subsequently proved to be correct”, said Paul Niven, manager of F&C Investment Trust. Speculation led to over-exuberance in the stock market during the dotcom boom, which detached stock prices from business fundamentals.

But Niven feels that “compared to the late 1990s, the adoption rates for generative AI have been quicker than during the dotcom boom and will, arguably, have a bigger overall economic impact in numerous ways”.

His assessment of the current market is that “while valuations are high and we are not complacent about the risks, I do not believe that we have an AI bubble at present”.

Could the AI boom diversify?

One of the alternative hypotheses about the trajectory of the AI boom is that, rather than continuing to see returns limited to mostly just the AI megacaps, it could diversify into broader sectors.

“Thinking about the new AI paradigm, two seams stand out,” says Gary Robinson, co-manager of Baillie Gifford US Growth Trust. “First, the application layer: businesses turning frontier models into real services for consumers and enterprises.” He highlights the Baillie Gifford US Growth Trust holdings such as DoorDash (NASDAQ:DASH), Shopify (NASDAQ:SHOP) and Roblox (NYSE:RBLX) that fall into this category.

“Second, the enablers: networks, data and security that form the new stack.” AI-powered cybersecurity stock Cloudflare (NYSE:NET) taps into this theme.

“So far, UK-listed firms have largely missed out on the AI party,” said Kenneth Lamont, principal at Morningstar. “Unlike the US Magnificent Seven, which are pouring billions into AI infrastructure, no UK stock can be considered a pure AI play.”

Lamont highlights AstraZeneca (LON:AZN) as one of the most-frequently held US-listed stock in global AI funds, followed by RELX (LON:REL), Experian (LON:EXPN) and LSEG (LON:LSEG).

“These are all AI applicators – businesses using AI to improve their models, such as AstraZeneca in drug discovery – rather than core AI providers. In the end, it may be these applicators that have the last laugh: by harnessing increasingly powerful and low-cost AI tools to boost margins with little risk, they could prove to be the real long-term winners,” says Lamont.

Protecting against the AI bubble bursting

You are likely more exposed to any potential bursting of the AI megacap bubble than you think. Most of your portfolio, whether that’s investments in a stocks and shares ISA or your pension, will likely comprise index funds that track the global stock market. If sentiment towards these stocks dips your portfolio could take a hit, at least in the short run. This concentration risk is one of the drawbacks of passive investing.

You can reduce your concentration risk by buying equal-weight funds, which will cap every holding at a certain weighting of the portfolio. Lamont highlights Xtrackers S&P 500 Equal Weight UCITS ETF (LON:XDWE) as an example.

Or alternatively, you could buy a fund that excludes the megacaps altogether, such as Amundi MSCI USA Ex Mega Cap UCITS ETF (LON:XMGA).

Investors that want to retain some exposure to AI without being over-exposed to the stretched valuations of the megacaps could look for value opportunities in AI and its broader ecosystem, such as the suppliers to companies like Nvidia and TSMC.

Rob Morgan, chief investment analyst at Charles Stanley, emphasises that diversification is the key to protecting your portfolio against market volatility.

“Owning different sectors, geographies and types of asset can help reduce risk and ensure you are not overly reliant on one area or on a particular set of circumstances,” he said. “In today’s context, investors should ensure their portfolios aren’t completely dominated by the theme of big tech and AI, including tracker funds heavily skewed towards large US stocks.”

Morgan recommends bonds as one offset to big tech concentration as well as defensively-minded global equity funds, or equity funds that target resilient, dividend-paying stocks.

“For instance, JO Hambro Global Opportunities offers a blend of offense and defence with the managers focused on quality and value. Zero exposure to Nvidia, Apple, Amazon, Meta, Broadcom and TSMC make the fund a more defensively minded diversifier to a global tracker,” he said. “Meanwhile, Trojan Global Income focuses on quality and resilience as a priority with a preference for businesses with minimal earnings volatility and low capital intensity.”