Artificial Intelligence
AI investment opportunities aren’t limited to tech
AI has become a major investment theme, driving the technology sector to new heights on the stock market.
Beyond the technology sphere, what other sectors are affected by this AI tidal wave? How is AI spreading to the rest of the economy? Answers with Guillaume Uettwiller, thematic manager at CPRAM, a recognized expert in thematic management.
Published on 01 July 2024
How do you define artificial intelligence?
Artificial intelligence is the ability of a machine or computer to simulate human intelligence to perform tasks and/or make decisions. In short, AI today is capable of reasoning, thinking, making decisions and even learning. It is a technology that is booming, which has been brought to the forefront with ChatGPT since November 2022, but which has been in our lives for several decades.
When you use your mobile phone to unlock for example via facial recognition, it is AI. When you look for the fastest route in traffic jams, it is AI. When you search directly on Amazon, you are using artificial intelligence. AI is already at the heart, and well anchored in our daily uses. But it will become omnipresent in the coming years because industrial use cases will develop. In finance obviously, but also in agriculture, industry or even health.
What are the factors driving the sector's performance?
2024 is a continuation of 2023. We are still in the construction phase of the artificial intelligence infrastructure that will support this development. Hyperscalers, which are public cloud providers like Google, Microsoft and Amazon, have significantly increased their investment capital expenditures.
They will peak at around 200 billion per year over the next few years, in 2024 and 2025, an increase of 40% compared to last year. And this is what has driven up AI values. We see these investments starting to trickle down into the ecosystem, into the AI value chain.
This ranges from semiconductors with equipment manufacturers, designs or even foundries, but also in communication networks or even computer hardware linked to servers, typically racks. And this is all what contributed to the group's very strong performance at the start of the year.
AI a phenomenon only linked to technological values?
We are entering a new technological super-cycle, so it is obviously the tech sector that is concerned first, but obviously, the construction of this infrastructure, that is to say these 200 billion that will trickle down, offers opportunities in other sectors of activity.
Typically, we take a concrete example, accelerated data centers consume 7 times more energy than a traditional data center, and so inevitably, this creates opportunities in renewable energies, particularly solar, but also in the modernization of the electricity grid, which will be a problem to accommodate data centers that can make a gigawatt of power, that is to say ten times more than the size of the data centers that we are currently building.
Obviously, there will be industrial use cases that will emerge in the coming years, whether in robotics or in health. We are already seeing pharmaceutical labs using generative AI to simulate protein structure and simulate interaction in medical research.
Why this gap between semiconductor manufacturers and software publishers?
Indeed, we had a fairly strong correlation between these two sectors in 2023, due to the fact that investors expected rapid monetization of new features related to artificial intelligence.
In 2024, the picture is a little more nuanced, I would say. We see that monetization is slower, except at Microsoft and ServiceNow, which have quantified the contribution of AI in terms of revenue. The other players who had benefited from a positive announcement effect were rather punished if it did not follow. The macroeconomic environment also remains difficult and we see that it weighs on the choice of companies and cost control.
And we also observe that the market remains rational since investors only react to one thing, namely positive or negative revisions of fundamentals.