Artificial intelligence revenues are still far from justifying the investment obsession

Digest opened free editor

Rola Khaleda, FT editor, chooses her favorite stories in this weekly newsletter.

A year ago, the Truc Motory Intelligence obsessed, which sweeps the Silicon Wall Street Valley, faced a serious examination of reality.

In a large scale quoted NoteJim Kovilo, head of stock research, Jim Kovilo, asked whether companies that plan to pour $ 1 billion in building obstetric intelligence would witness money. A partner in the Capital Capital Capital, at the same time, estimated Technology companies need to generate additional $ 600 billion to justify additional capital spending in 2024 alone – about six times more than they would have been produced.

Warnings have helped to raise the first real test of investment morale since the launch of the electrified Chatgpt from the industry. The revenues of the final customers who were supposed to benefit from this new technology were minimal. Where were the “killer applications” of artificial intelligence? This led to a summer of anxiety for technology investors.

A year later, the leading artificial intelligence shares were with another volatile swing. Get rid of fears that started with low -cost artificial intelligence models created by Dibsic in China, Nafidia brisk To get a new standard level this week, with a profit of about $ 1.5 trillion of market value from its lowest level in April. Microsoft has also recorded a 1TN market bounce in less than three months.

Nevertheless, what is noticeable is the extent of a little change in the wider revenue expectations in artificial intelligence since warnings a year ago. Hope (noise) is strong as it was always, but it is still difficult to know where the returns will come to justify the huge capitalist spending on artificial intelligence, at least in the short term.

On the cost side, the effects of artificial intelligence obsession are very clear. The four technology companies that lead these fees-Alphabet, Amazon, Meta and Microsoft-increased by about two-thirds of the capital spending in 2024. With the start of this year, they were Planning to enhance CAPEX by another $ 75 billion.

There is an additional escalation during the rest of the contract that is baked in expectations. Bank of America predicts that for the technology industry as a whole, spending on data centers will jump from $ 333 billion last year to about one dollar in 2030. By the end of the period, 83 percent of the money will go to the prosecution’s investments.

On the aspect of revenue from the equation, at the same time, some artificial intelligence leaders began obtaining an increase in the percentage of business – but additional revenues are calculated in tens of billions instead of hundreds.

Early this year, Microsoft said that its annual revenue from artificial intelligence increased by 175 percent to $ 13 billion. This is still about 5 percent of the total revenue that is expected to produce this year. The average revenue in Openai from subscriptions, its main income source, Only 10 billion dollarsDouble from the end of last year. The increase rates are noticeable, but the absolute numbers are still pale compared to Capex.

There were also signs of an explosion in the use of Chatbot a year ago, as Openai converted almost overnight to an unlikely consumer technical company. But while large numbers of people are now using AI Chatbots, the work is still small. About 3 percent pay the price of the artificial intelligence service they use, which leads to annual revenues of about $ 12 billion, according to a survey of 5,000 American adults from Menlo Ventures.

Meanwhile, the business scientist is still looking for the uses of Amnesty International to justify serious spending. The technology has been fed at least New generation One of the highly developed startups, especially in coding, the first field of “knowledge work” is seriously disrupted. But it has not yet caused the growth of revenue for the largest software companies, which it is best to place to bring artificial intelligence to the business world in the form of a new generation of Acting applications.

The first wave of spectra and aids of artificial intelligence did not do much to change the practical life. Hope It has now turned To agents – tools capable of automating individual tasks, or even entire work operations. According to Macinzi, agents Be a creativity The value of dangerous works by automating complex and important operations. But consultants also warn that this will require re -thinking about the entire commercial operations.

Convincing its customers to overcome many experimental projects in Genai, which publishes the world of companies, will serve as a heavy elevator for the technology industry. This does not mean that, in the long run, the Toulyi AI has no chance to achieve the type of transformation in the practical life that its reinforcement calls. However, at the present time, the gap between capitalist spending and revenues did not appear on a little narrowing.

[email protected]