AI investing: where the next winners could emerge

The way we think about investing in AI is changing as fast as the technology itself. The right questions to ask are evolving. Is the AI revolution real? Obviously. Will it deliver returns to match the high cost of building it out? Maybe. Who will capture the economic value? That’s harder.

New technologies often trace a similar pattern. Railways, electricity, cars, the internet and now AI have had three phases. First, an innovative technology. Next, the infrastructure to deliver that technology to a mass market. Only then do the real beneficiaries emerge.

In each case, the migration of value has been determined by a shift from scarcity to abundance. The right question to ask now is what happens and who benefits when what is scarce today becomes abundant tomorrow. It is as true of AI as it was of the previous episodes.

• Get personalised advice on savings, investing, retirement or pensions from Fidelity’s financial advisory service

Take the railway boom. The scarce resource 200 years ago was rapid and cheap transport. This became abundant thanks to the build out of the railway network and the early benefit was captured by railway promoters and operators. But the ultimate value quickly migrated to manufacturers and farmers, who could move their goods to big new markets, and the retailers who could sell them.

Electricity is perhaps an even better analogy for AI. Both are what economists call a ‘general purpose’ technology. Again, the real beneficiaries turned out not to be the generators and distributors of a commodified product but the businesses that used electricity to redesign the way they operated.

Replacing a big steam engine with a big electric motor in an unchanged manufacturing process was not the key breakthrough. The productivity gains came from putting small electric motors next to individual machines and redesigning the factory around the new technology. This is why the benefits of electrification took decades to be felt.

• Get the latest market updates, ISA news and pension insights by email

In the same way, it will only be when everyone’s computer is running Claude or Perplexity, and we’ve all worked out how to use it, that AI will be transformative. When that happens, the gains accruing to the chip-makers and the builders of the data centres won’t disappear. But an increasing share of the economic value might migrate to the businesses and individuals using cheap and abundant intelligence.

The invention of the internal combustion engine is the best example of the unpredictable benefits of a new technology. Again, there were first-order winners: the car makers like Ford and General Motors, obviously; then the oil companies, tyre manufacturers, steel producers and road builders. But the introduction of ubiquitous personal mobility changed forever where people could live, shop and work. That led to suburbs and the growth of leisure and tourism, and ultimately a mass consumer economy.

So, another pertinent question is: what becomes possible when everyone has AI?

Bringing us almost up to date, this is what happened with the internet revolution a generation ago. Again, the initial focus was on the infrastructure. But this missed the real significance. The investment error was assuming that booming demand would primarily lead to attractive returns for the companies supplying the internet.

As with all the previous technologies, the infrastructure build-out made bandwidth, computing power and connectivity cheap. And that was the key to shifting the economic opportunity from the suppliers to the users of the technology. Cheap connectivity created one wave of winners; ubiquitous mobile computing subsequently created another.

Some of the initial beneficiaries survived this migration of value. Amazon is the obvious example. Some of the second wave have thrived. Think Google and Meta. But others were hard to spot in the early stages. The inventors of the internet did not have takeaway pizza in mind, but Domino’s became a digital ordering and logistics business that just happened to have fast food as its product.

Perhaps the best analogy of all for AI is what happened when GPS stopped being a scarce and expensive military capability and became an essentially free infrastructure embedded in everyone’s phone. Who benefited? Not the makers of GPS equipment. It was Uber, Deliveroo, Google Maps, Strava.

What was once scarce disappeared into the background and became a utility. Substitute intelligence for location and investing in the AI theme suddenly becomes very interesting. What happens when intelligence itself becomes an abundant resource?

I think about the AI value chain as having three layers - innovators, enablers, and adopters. And if you want you can invest in each of these simply through thematic ETFs. iShares covers the waterfront with: AI Innovation Active, AI Infrastructure, and AI Adopters & Applications.

What’s most interesting about these three funds is that the further away you travel from the hype at the top of the chain towards the more mundane companies enabling and using AI, the less you have to pay to participate in the theme. The innovation, infrastructure and adopters funds trade on 52 times earnings, 44 times and 27 times respectively.

• Read: Is the AI boom hiding an earnings bubble?

The biggest risk for today’s AI winners is not that AI fails but that it is so successful that the scarcity they can sell so profitably today becomes an abundant and cheap commodity. If I were putting together a portfolio of innovators, enablers and adopters, it would not be an equal weighted basket. The cheaper and more durable end-users of AI is where I would focus.

The winners in this category may well be those who use what is abundant to benefit from what remains scarce - that’s likely to be some combination of proprietary data, trusted brands, customer relationships, human interpretation and judgement or even physical assets. Things that AI cannot commoditise.

This article was originally published in The Telegraph.

Got a burning question you want to ask? Why not drop us a line. Click here to ask your question.

• Read: 5 stocks to watch in September
• Read: Why are gold and bitcoin rising together?
• Read: Tracker funds that avoid the Mag 7, chip stocks and SpaceX

Important information - investors should note that the views expressed may no longer be current and may have already been acted upon. Reference to specific securities should not be construed as a recommendation to buy or sell these securities and is included for the purposes of illustration only. Overseas investments will be affected by movements in currency exchange rates. This information is not a personal recommendation for any particular investment. If you are unsure about the suitability of an investment you should speak to one of Fidelity’s advisers or an authorised financial adviser of your choice.