The Viewpoint: Is AI a bubble waiting to burst?

All-time highs

Equity markets continue to hit fresh all-time highs despite concerns over geopolitics, valuations, inflation, economic growth and many other perennial worries. Of course, over time, equity markets do tend to rise as they benefit from innovation and economic growth, both of which have increased since markets existed. This suggests that all-time highs are a sign of health, not concern, however equity investing is prone to over exuberance at times and the reasons for new highs are always worth understanding.

Many of the questions facing markets today concern artificial intelligence (AI), notably whether this is a bubble. As the saying goes, the only difference between a bull market and a bubble is if you are invested in it or not. Those with AI exposure in their funds argue it is a bull market, those without that it is a bubble!

Before we discuss AI further, there are other reasons global markets are touching all-time highs. Many governments are running fiscal deficits, which are stimulative to economic growth. Low oil prices, which are half the levels they were three years ago, are also acting as a kind of tax cut for consumers. Finally falling interest rates are another reason to be optimistic about economic growth. It seems to me that risk investors should be concerned about economic growth being stronger than expected, not weaker.

A further tailwind for markets is investment in AI, which now accounts for around 40% of US GDP growth this year. This investment, into the data centres and ancillary products & services, is of a scale rarely seen before. Comparisons have been made with the build-out of rail infrastructure over 100 years ago, and more recently the build-out of fibre in the early 2000s to support the internet. Inevitably, these comparisons, and its materiality to economic growth, have created concerns that this is a bubble and that if it bursts, it would have negative consequences for economies and markets.

As I have noted before, no one has privileged access to the future, therefore no one definitively knows if we are in an AI bubble or not. The arguments we have come from the lack of revenues and profits which can be identified from AI. OpenAI, the parent of ChatGPT, has 800 million subscribers but only a small percentage of these are believed to be fee-paying. The revenues from developing AI models, building applications based on it, and running the AI infrastructure such as data centres have not yet justified the huge investment in this area.

Optimists will point to a similar narrative around the internet in the late 1990s, where short-term overbuild of infrastructure was compensated for by long-term demand and applications which changed society in profound ways. It is true to say that we are unlikely to be able to comprehend how pervasive AI will become and the value it will add to our lives. In the same way that the internet created huge value for investors and companies, AI may do the same.

What is interesting about these counter arguments is they do not need to be mutually exclusive. It is possible to believe we are in a bubble and also that AI is the most important trend in society today.  From when the internet bubble burst in January 2000, to when the fall-out ended in September 2002, Nasdaq (the US index of technology shares) fell 75%, since when it has increased by over 1800%! Extremes of optimism and pessimism are the nature of technology investing.

As someone who started their career in the latter stages of the internet bubble and subsequent bust, there are parallels today, but these are not yet compelling. There is a huge amount of investment going into areas where the returns on it are far from clear. Equally, share price moves connected to anything AI-related are reaching extreme levels. That said, in my view the level of AI exuberance is not yet at internet levels, and the main AI investment options are much more established businesses than those internet start-ups back in the late 1990s. Bubbles burst when everyone believes in a new paradigm (think the US housing market in 2007, sovereign debt in 2021) not when there is widespread scepticism as there is today.

In the end, business fundamentals will determine the future of AI as a driver for markets and companies. If use cases can be found which enhance productivity and build valuable products and services, then revenues associate with AI will justify current and future investment. If not, then, like the overbuild of fibre relating to the internet boom, there will be pain ahead. For now, I think the best approach is to keep an open mind.

Is China investable again?

One feature of markets today is that there are a range of ways to make money. Although much of the attention is on AI, which inevitably leads to the US market, almost every equity market globally has performed well this year. One example of this is China, whose equity market is up 20% year to date, and nearly 50% since its lows in September 2024, when many commentators declared that China was uninvestable.

Like the mantra of the end of US exceptionalism earlier this year, when a sweeping generalisation is made into a catchy line, it is usually time to invest in the opposite direction. Since the end of US exceptionalism was declared, US markets have performed strongly. If it’s in the paper, it’s in the price is a good rule of thumb.

The Chinese stock market has performed well due to an increasing understanding that it has taken leadership in several key industrial areas, from cars to renewable energy and robotics. This has been a consequence of President Trump’s first term when his aggressive approach to China convinced it that they must become self sufficient and not rely on a US partner in the future. As a result, China has spent the last decade investing in infrastructure and industry in a way which the US is now trying to do itself. This is one reason why there has been no tariff deal between the US and China: they have equal leverage against each other.

This Chinese renaissance has happened while many institutions in the US appear to be under threat. Independence of the Federal Reserve, free press, and the judicial system are things historically more associated with China than the US. When will investors ask is the US investable? We doubt the US will become uninvestable, however markets move at the margins, and the US degrading while China is improving is a notable trend.

In the end the broadening out of sources of investment performance is a good thing. Whilst AI takes most of the headlines currently, there are bull markets developing in other major economic regions with China being a good example. Although AI will have an important role to play in the future path of market, it is important to remember there are other reasons to be positive too.

For professional investors only.  This material is not suitable for a retail audience. Capital at risk. This is a financial promotion and is not investment advice. Past performance is not a guide to future performance. Reference to any security is for information purposes only and should not be considered a recommendation to buy or sell.

The value of investments and any income from them may go down as well as up and is not guaranteed. Investors may not get back the amount invested. Portfolio characteristics and holdings are subject to change without notice. The views expressed are those of the author at the date of publication unless otherwise indicated, which are subject to change. Forward looking statements are subject to certain risks and uncertainties. Actual outcomes may be materially different from those expressed or implied.