Jason Hollands: Is artificial intelligence the next investment bubble?

In fact, while the S&P 500 Index of large US-listed companies is seemingly up by 9% year-to-date, its gains are entirely down to technology and online businesses. Absent the companies in the technology and related communications sectors, which together make up 37% of this benchmark, the 418 remaining companies in the S&P 500 outside of these sectors are collectively down -2.4%.

Typically an environment of rising borrowing costs, as we are in now, is a negative headwind for growth sectors like technology, but what has turbo-charged their share prices in recent months is excitement over Artificial Intelligence (AI) – the development of machines that can rapidly process vast amounts of data in real-time, learn, adapt and take decisions.

While AI has been in development for some time and is already been applied by some businesses in areas like managing supply chain logistics, fraud detection, as well as digital voice assistants like Amazon Alexa and Apple’s Siri, interest in AI and its potential has exploded since the launch of the ChatGPT app in November 2022.

The free-to-use AI “chatbot” tool enables users to have human like conversations and can be used to generate content such as articles. The wave of interest has been unprecedented with ChatGPT reaching one million users in five days, smashing the records of other apps in gaining their first million users such as Spotify (150 days) and Facebook (300 days).

The possibilities of the development of such disruptive technologies are enormous, perhaps even more so than the arrival of the internet which has undoubtedly had a major impact on how we live our lives.

The potential application of AI is across many areas of life including improving productivity, scientific research, resource management and even culture with the creation of art and music.

Goldman Sachs estimate that generative AI tools – a branch of AI that creates content like articles, communications and images, could alone drive a 7% increase in global GDP and lift productivity growth by 1.5% over a 10-year period.

Alongside other applications, such as powering factories, where robots function around the clock don’t strike, and medical discovery, AI could be a driver of the next industrial revolution.

But it has also provoked considerable debate about the risks, ethics and the need for regulation. Tesla founder and Twitter owner Elon Musk earlier this year co-signed a letter warning about the risks of developing “non-human minds that eventually outnumber, outsmart, obsolete and replace us". And this week over 300 leading figures in the technology sector issued an almost apocalyptic warning AI could lead to the extinction of humanity.

Dangers identified include driving an exponential increase in the spread of misinformation, enabling authoritarian regimes to enforce surveillance and censorship, and the development of new weapons.

The huge increases in productivity from AI will come at a price: the displacement of millions of jobs currently performed by humans. This is not just factory workers, but white-collar jobs are at risk too.

Goldman Sachs estimates two-thirds of occupations could be partially automated and globally 300 million jobs could be displaced by automation. However, new roles will emerge too as they have done in previous periods of industrial innovation, with large firms already starting to step up recruitment in AI-related roles.

As investors pour cash into AI-related investments, it is wise to heed the lessons of the dot-com bubble at the turn of the 21st century, the eventual burst of which led to huge destruction of capital.

While some businesses both survived this crash and went on to thrive, others have sunk without trace as their lack of actual profits and flaky business models were exposed.

Inevitably all manner of large companies are starting to talk about how they are using AI – as many similarly did about their online presence during the dot com bubble. Increased noise around AI from corporations will make it harder for investors to identify the more compelling plays.

It is often said that in the Great American Gold Rush, the guys selling picks and shovels made more money than the average prospector and this may well be true when it comes to investing in AI.

Identifying which edgy new AI developers will prove winners over the long run is like picking a needle in a haystack. What we do know is that the computing power essential to developing AI is going to create enormous demand for micro-chips, benefiting giants such as NVIDIA, ASML and TSMC.

As AI also relies on processing vast amounts of quality data, online platforms like Alphabet, the owner of Google, and Meta, the owner of Facebook, will be beneficiaries.

Investors have already caught on to this, with NVIDIA’s share rocketing 178% since the start of the year with its market valuation briefly reaching $1 trillion, placing it among the five largest companies on the globe. Its incredible ascent included a one-day rise of 24%, following a recent earnings call where the company outlined the enormous demand for its chips due to firms developing their AI capabilities, resetting investor expectations.

In periods of market euphoria like this, it is important not to throw caution to the wind, as the ballooning valuations of perceived AI beneficiaries are pricing in an enormous amount of optimism about future growth in profits.

The 68 AI-linked companies in the Robo Global Artificial Intelligence Index, are in aggregate valued at a very stretching 42 times their forecast earnings for the next year, way higher than the FTSE 100 on 10.5 times.

But when one pocket of the market is attracting large amounts of hot money, valuation bubbles can run on for some time before bursting. At the peak of the dot-com bubble, technology shares traded at 70 times their earnings before crashing back to earth, so at 42 times AI stocks could have some way to go to match this.

In conclusion, AI is game-changing technology which we regard as an investment megatrend that should not be ignored. But excitement and enthusiasm does need to tempered by understanding the risks of the more speculative plays and it is vital to keep a careful eye on valuations if they become disconnected from reality.

Jason Hollands is managing director at wealth manager Evelyn Partners which operates form offices in Glasgow, Edinburgh and Aberdeen.