Will AI Investment Continue Skyrocketing?

The magnitude of discussion on artificial intelligence (AI) has possibly reached its peak. The AI boom, AI investment etc. have become the hot topics among the industry leaders, financial analysts, economists and even policymakers. It is true that the AI sector has created mammoth investment opportunities not only in the USA but also in other parts of the world.

Specifically, building data centres has created massive investment prospects, although very few countries have been able to seize the opportunity. Besides the USA, Canada and India are two countries that have been able to capture a good chunk of AI investment. India has received the commitment from US-based tech giants to invest close to a trillion dollars for building data centres. Thus, AI investment has taken a massive turn with the promise of trillions of dollars of investment in the coming years. 

Reportedly, the combined capital spending in the AI sector made by four tech giants, which include Google, Microsoft, Amazon and Meta Platforms, has already exceeded $168 billion during the first half of the current year ending in June, which is about 74% higher than the previous year. This massive investment in AI infrastructure has been allegedly impacting both the free cash flow and stock prices of these tech giants. Because of continuously massive investment in the AI sector, these tech giants are termed ‘hyperscale’ in the AI arena. Economists, financial analysts, and industry experts are sceptical and sort of worried about such massive investment, particularly in one sector. However, anticipated revenue of $1.25 billion per month from one AI company, Anthropic’s AI capacity-building investment, has been working as rationality and reasoning behind massive AI investment made by hyperscale companies. 

Not only hyperscale companies but also many other AI companies have been in the race of significantly investing in this sector. Elon Musk’s SpaceX is one such example, because SpaceX made a $12.7 billion capital expenditure last year on building its AI infrastructure with an additional $37 billion investment to be made this year for AI capacity building, as estimated by industry experts. Nevertheless, the magnitude of investment in AI infrastructure is mostly confined within a few hyperscale companies. However, the performance and outcome from the massive investment in the AI sector are not at the same level for all hyperscale companies. As an example, Meta Platforms is one of the hyperscale companies, but they have been reportedly lagging behind other competitors, viz., Amazon, Microsoft and Google, which have been selling their cloud services to businesses for the last decade.

One industry expert has opined that the scale of Meta’s network is already its rival’s cloud provider footprint. The expert further estimates that Meta has about 20 gigawatts (GW) of computing capacity with an additional 14 GW coming online over the next few years. So, if Meta starts renting out their capacity, it will indicate that the company has also built overcapacity. In this connection, Meta’s CEO Mark Zuckerberg has said in the company’s annual shareholder meeting.

Although Meta is considered a smaller company in terms of AI infrastructure compared to other tech giants, Amazon, Microsoft and Google, this company has recently been highly ambitious in AI investments. Meta has established a division which is called the Superintelligence Lab, with an objective of pushing for the social media network to be the first to develop a supercharged form of AI. It is learned that Meta is planning to raise investment by taking over half of its annual revenue this year on building AI infrastructure. Many analysts and industry experts have expressed their doubts that Meta will pause or reverse their investment in AI infrastructure. One industry expert has opined that Meta is not stepping away from the AI race; it is turning early, aggressive capacity commitments into a strategic value creation option. The frantic race among the hyperscale companies in AI investment has created a mixed reaction among the investors, as one group is found to be very optimistic about a positive outcome, while the other group is found to be very pessimistic about the long-term consequences of the huge investment in one sector.

The real situation of massive AI investment may be revealed in the coming round of earnings reports to be presented by the hyperscale companies, which may indicate in which direction AI investment euphoria will move. But the industry experts opine that the AI race will continue with massively investing in this sector for the coming years, whatever the earnings may be. Because, based on the own projections of four hyperscale companies, combined capital expenditure in AI by Google, Amazon, Meta and Microsoft is expected to reach $710 billion this year and $1 trillion in 2027.   

Now a question arises: whether renting out overcapacity of AI infrastructure will be treated as a short-term offset to continuously massive investment or be the cause of economic recession, the answer to which still remains to be seen. People of the related field, government authorities and even policymakers do not appear to be looking far and thinking about the consequences. A euphoria has been sweeping across the AI industry, where a craze of investment has been going on without any sign of a pause. It seems that AI investment will continue skyrocketing unabated.