DeepSeek seems like a long time ago now. In fact, it has been less than a year since the introduction of China’s AI champion was being talked about as a potential ‘Sputnik Moment’ that would see the US lose its technological edge to a communist state.
Some $1trn was wiped off the Nasdaq in a matter of days. Nvidia alone fell by close to 40% between January 23 and April 4, with the ‘Liberation Day’ tariff announcements giving it another kick while it was down. The view at the time was that DeepSeek, which was developed at a much lower cost than its US competitors, would kill investment in AI.
We felt this was the wrong way of looking at it, and we have been proved right. Expected AI capex for 2028 has since increased from $250bn to $500bn. We think it will be closer to $1trn. Nvidia is up by more than 100% since bottoming out in April.
Exceeding expectations
Exceeding expectations has been a common theme since we first invested in Nvidia in 2023. This is one of the reasons why we are sceptical about claims of another dot-com bubble – and there are many others.
Each of the Magnificent Seven has a set of characteristics that make it attractive to investors. All these companies have consistently delivered some of the highest earnings growth in the market. They have dominant positions in the areas in which they operate. And, for all the talk of excessive valuations, they have average free cashflow yields of 16%.
Last quarter was a strong one for the tech giants. However, rather than surging on the good news, some actually fell – making them cheaper as they grew faster. This is not what ‘irrational exuberance’ looks like.
Not a bubble in the traditional sense
Yet here we would sound a note of caution. We do not think the US market has entered a bubble of the type that peaked around the turn of the millennium. We instead think the bubble – if there is one – will take a different form.
AI presents an existential threat as well as an opportunity to the business models of the tech giants. Most currently operate in effective monopolies, but AI means every category in which they compete is now up for grabs – whether that be e-commerce, digital ad sales or even social media posts.
Greater competition will bring down prices for customers, while the cost of delivering these services will rise as providers are forced to spend more on AI accelerators.
As a result, we think any bubble for the tech giants will be in the cashflows they generate. These will eventually fall back from their elevated levels towards the market average, with valuations potentially going the same way.
How to make money from the AI boom
Once you have taken this on board, you can begin to think about how to make money from the AI boom – because there is still plenty to be made.
Inevitably, there will be winners and losers among the tech giants. But we do not yet know who they will be. We want to retain an open mind so that we can act accordingly when the evidence presents itself.
One thing we are fairly certain about, however, is that the huge sums already allocated to AI are going to be spent – and then some. The heads of the hyperscalers – the providers of massive cloud-computing services – are terrified of being left behind, and one of the main determinants of who wins and loses will be who has secured enough chips.
The old cliché about the ‘picks and shovels’ approach to making money from a bubble therefore first leads us to Nvidia. The world is going to need every chip the company can produce for at least the next three years.
AI’s reliance on energy
Another potential bottleneck for these businesses is securing enough energy for their data centres. With nuclear power unlikely to be the answer for at least five years, the immediate future likely lies in a mix of natural gas, solar and fuel cells.
The beneficiaries should go all the way down to boring old regulated utilities. We are used to these growing at a fairly unexciting 5% a year, but many have now raised their forecasts to 9% – a figure pretty much nailed on following negotiations with the utility commission
AI’s blue-collar boost
Another consideration with data centres is the issue of who will build them. The supply of skilled labour could be as much of a bottleneck as the supply of energy or microchips.
We recently met the chief financial officer of a holding that is benefiting from the build-out of AI centres. Previously, his business’s fortunes were closely tied to economic growth. Even in the good times a downturn was never too far away. Decades of this experience typically make chief financial officers averse to hyperbole. We were struck by his observation that, because of the expected size of the AI build-out and the contracts for other industries getting squeezed out to be built later – he did not expect a blue-collar recession in his lifetime.
There is plenty of excitement around AI and its potential to transform and disrupt even the most advanced industries. This time around, though, taking a ‘picks and shovels’ approach to a stock market boom could well mean investing in the people that actually use picks and shovels.
Cormac Weldon is Head of US Equities at Artemis
