The US S&P 500 index has gained 18% this year. Many money managers have been caught out. Their defensive focus on rising inflation and interest rates left them blindsided by big gains in Artificial Intelligence (AI) stocks. There’s no question that long-term investors should be exposed to AI in their portfolios. It is no fad and will drive efficiencies and profitability across diverse industries for years to come in ways we cannot yet imagine.
But should investors buy the types of Big Tech stocks that are capitalising on AI now, but look expensive because they have spearheaded the market’s advance this year? Or should they instead try to identify up-and-coming businesses with a future in the field hitherto overlooked by other investors?
The idea of finding hidden gems is, of course, appealing and there will be many as yet unseen winners from AI. And there is no shortage of ideas about the ways in which it could be applied and even transform society. They range from cancer research to robotics to deep-space exploration.
Subscribe to 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
But all AI’s potential applications rely on processing vast amounts of data extremely quickly, and just any microchip won’t do – you need the fastest. The favourites are graphics chips that can work on mulicetiple tasks quickly enough to produce the often visually stunning video-game content, but they are scarce.
A chip supply squeeze
The world faces an overall shortage of the microchips needed to meet the strong processing demand induced by AI. New entrants looking to build businesses in the sector simply do not have the resources to achieve the necessary scale with a product that is already difficult and expensive to secure. So they have to rely on effectively renting use and access to computing power from the big players.
This means the practical and financial dynamics of the emerging AI industry favour the incumbent Big Tech players, and that won’t change soon. These businesses can provide chips and access to vast computing networks with powerful processing capacity, and so this is where the money is most likely to be made, at least for now. These big players are miles ahead of everyone else when it comes to having the equipment that matters.
This means that despite the strong gains so far this year, they remains promising long-term plays. Towards the end of June, CNBC, the US business television channel, surveyed about 400 professional investors, from chief investment officers to fund managers. When asked what they think is the best way to invest in AI, nearly half said via Big Tech. Only 16% suggested going for up-and-coming stocks instead, with the rest sitting on the sidelines.
The dominant chipmaker
Drilling down to which individual stocks they like most for investing in AI, microchip-maker Nvidia scored 50% and Microsoft 28%. Other dominant tech players lagged far behind: Alphabet, which owns Google, and Amazon, scored 11% and 6% respectively. Nvidia is the US microprocessor business that started up in the 1990s and has become the most influential and innovative chip designer in the world.
While the launch of ChatGPT fired the imagination of the broader public about the possibilities of AI, it was a very bullish profit announcement from Nvidia in May that put AI squarely in the cross hairs of growth investors everywhere.
Having achieved sales of $7.2bn in the first quarter, the group said it expected $11bn just three months later, a 53% leap. We suggested considering the shares before the announcement, and they have now risen by about 50%. This has meant that despite already being one of the biggest companies in the world, Nvidia is now one of the few companies valued at more than $1trn, entering a pantheon alongside the likes of Microsoft, Alphabet, Apple and Amazon.
Nvidia is a hold even though it has performed strongly because we believe in a long-term investment approach: AI is a multi-year trend. Trying to trade the shares for further short-term gains could work but is unduly risky. Its high-speed chips are essential and highly sought after for building the infrastructure needed to satisfy the take-up of AI, and it supplies 80% of the market. Demand will keep Nvidia chip pricing strong and sales volumes high.
Unsurprisingly, the company is valued significantly higher than the market as a whole. It can be justified, but the company needs to deliver. There are some 50 analysts researching the company and 43 of them expect further outperformance. We are in one of the most exciting and transformative shifts in technology in world history, and owners Nvidia are at the centre of it.
Microsoft steals a march on rivals
Software and computing giant Microsoft is also at the heart of the AI revolution and thus likely to be a leading beneficiary as the mega trend develops over coming years. It is set to report quarterly results in a week or so when investors are likely to hear plenty of plans and developments around AI which could kindle market excitement.
And when it comes to the numbers, Microsoft has a pretty good track record of outstripping investors’ expectations. For the second quarter of 2023 the estimates are earnings growth of 14% to $2.55 a share on revenues up by roughly 7% to $55.4bn.
Microsoft doesn’t design microchips like Nvidia. It provides software applications such as Microsoft Office, and has a data storage arm called Azure. Anyone can pay to access it remotely via the cloud, saving a fortune on kitting out and overseeing their own data infrastructure.
It’s also building a gaming entertainment business, which is a significant growth area, and after much regulatory resistance, is close to completing the acquisition of Activision Blizzard (see page 6), one of the world’s biggest video-game publishers.
Microsoft has, over recent years, reportedly spent billions of dollars stealing a march over rivals by quietly building a stake in OpenAI, the developer and owner of ChatGPT, the AI conversational interface that has brought AI into popular use and consciousness. This places Microsoft at the heart of how AI will transform computing and how we use it. Programmers, software developers, business managers, investors, bankers and politicians are also involved in shaping the debate.
And of course OpenAI needs a company like Microsoft as it gives it vast computing and data- processing power, which is in short supply everywhere. As everyone turns to OpenAI to help get their AI work off the ground because they cannot build their own infrastructure owing to a lack of parts and other resources, Microsoft is in the background renting out the capacity to make it all happen.
On top of what is in effect a royalty on AI, the tech giant is busily integrating OpenAI into its own products, including its cloud offering, sales and programming applications, the Bing search engine and the Office suite. Only this week , it launched CoPilot, a new application embedded within Office, providing users with AI capabilities.
Has Alphabet fallen behind?
Then there is Google’s parent company Alphabet. It is also carving out a leading space in AI, although this year it has often been viewed as falling behind competitors despite working in the field below the radar for years.
The stock price is also affected by the fortunes of its economically sensitive Google search-based advertising revenues. At times, too, there has been an existential question mark over the business: if people can have their internet questions answered in a polite, succinct conversational style with Microsoft’s ChatGPT, why do they need the reams of web pages that Google increasingly seems to produce randomly, not to mention the creepy, targeted adverts?
The bouts of negative sentiment are probably an opportunity, again for the long-term. Alphabet has, in fact, made been making significant investments in AI over the years, too. The future outlook for traditional search engines and the advertising model that provides the bulk of Alphabet’s revenues is murky, and the prospect of difficulties on that front should not be lightly dismissed.
But, like Microsoft, Alphabet is at the centre of AI. It has also spent a lot of money building a cloud offering and is now number three in the sector after Amazon and Microsoft. This is helping to tilt the overall group revenue mix from away from advertising as it also caters to the need for processing power.
It has been consolidating its AI investments such as Deep Mind. The AI research across the group is being integrated into its broad range of products from YouTube through to Android, with a stream of new AI tools for its users launched recently. Meanwhile, Alphabet has a very strong balance sheet, and the stock trades on a valuation significantly lower that Nvidia and Microsoft.
Sticking with Big Tech should reward investors with what AI has to offer over the medium term. For now, the AI landscape resembles a shaken snow globe. No- one knows how it will settle. What we do know is that this nascent field needs computing power and we do not have enough to meet demand. This is where AI money is to be made until we get a clearer picture.
Stephen Connolly writes on business and finance and has worked in investment banking and asset management for nearly 30 years. (sc@plainmoney.co.uk)
Is property investment still as safe as houses? Why golden era could be over
What Santander’s takeover of TSB means for customers
