We don't yet know exactly what artificial intelligence (AI) will bring us in the future. But the technology is improving rapidly and is already causing two phenomena that we couldn't have anticipated three years ago. The first is the surprising prolongation of the economic expansion in the US despite the sharp rise in interest rates between 2022 and 2023. Investment in data centers, power plants, chips, and other AI-related goods accounts for a very significant portion of US GDP growth. Jason Furman, former director of the Council of Economic Advisers, states that it constitutes 92% of the growth in his country's aggregate demand in the first half of 2025. The second phenomenon is the stock market bubble associated with AI. The surge in the valuations of the five major tech companies—Apple, Microsoft, Meta, Alphabet, and Amazon—has propelled the US stock market to new highs this year, even though the international geopolitical environment has been extremely uncertain since the start of Trump's second term. As is always the case with bubbles, it's impossible to know when they will burst and what portion of the stock price increase is based on solid foundations. Do the current valuations reflect realistic expectations of the profits these tech giants will generate? Will they all achieve high returns, or only those that win the race?
We are undoubtedly witnessing a frenzied investment race. Companies, and especially their charismatic leaders, are searching for their Holy Grail: a technology that provides superhuman intelligence, capable of creating and dominating new markets and generating groundbreaking scientific advances for the progress of humanity, from curing cancer to generating energy from nothing. Sometimes it seems that humans believe, once again, that it's possible to create a perpetual motion machine.
One of the most interesting features of this AI investment race is that Apple, a traditionally highly innovative company, is investing far less than its competitors. One possible explanation is that its leaders are aware they are already behind the curve. The shortcomings of their Siri have been known for some time. Furthermore, it may be perfectly rational for Apple to prefer being a follower in this technology, rather than a pioneer. They may believe that winning the race could end up being a curse that weakens the winners. This could happen for two reasons.
The first is that they invest far more than necessary. And the second is that, once the race is over, the benefits obtained are lower than expected. The probability of overinvestment is high. The enormous sums being spent on data, computing power, and energy are based on the idea that simply increasing the resources invested will proportionally improve the quality of AI. For an economist, this means believing that technology exhibits endless constant returns to scale. Experience tells us that this never happens. In fact, AI is very powerful, but it also has significant shortcomings. Latest-generation models continue to make mistakes in simple problems (such as The Will Rogers Paradox), and this indicates that the path to superhuman intelligence is not just a matter of increasing scale.
As for profits, Apple is an excellently positioned company to assess this. The company is extraordinarily profitable, but the source of its profits is not a specific technology but rather creating its own ecosystem, which integrates technologically advanced products and services and is very difficult for its competitors to copy. Whoever wins the AI race must ensure that the advantage gained is not easily eroded. It will not be enough to arrive first if those who come after can imitate you.
It is not obvious, then, that the fierce investment competition among the big tech companies is rational. In part, it is also explained by their rivalry with China and national security issues. And an additional, not insignificant factor is the matter of pride. The sheer desire to win of the leaders of Silicon Valley. A desire that some would call a pathological competitive spirit. Like in the myth of Icarus, the great tycoons of the AI age risk believing themselves to be gods and flying too close to the sun.
