The artificial intelligence industry needs to earn $600 billion a year to pay for massive equipment costs.
Despite large-scale investments by high-tech giants in artificial intelligence infrastructure, the growth of artificial intelligence revenues has not yet materialized, indicating a significant gap in the value of the ecosystem for the end user. David Kahn, an analyst at Sequoia Capital, estimates that artificial intelligence companies will have to earn about $600 billion a year to pay for their artificial intelligence infrastructure, such as data centers.
Last year, Nvidia earned $47.5 billion in revenue from data center hardware (most of the hardware consists of graphics processors for artificial intelligence and HPC applications). Companies such as AWS, Google, Meta, Microsoft and many others have invested heavily in their artificial intelligence infrastructure for applications like OpenAI's ChatGPT in 2023. However, will they return these investments? David Kahn believes that this may mean that we are witnessing the growth of a financial bubble.
Kahn's math is relatively simple. First, it doubles Nvidia's forecast for operating revenue to cover the total costs of artificial intelligence data centers (GPUs make up half; the rest includes energy, buildings, and backup generators). It then doubles that amount again to provide 50% of gross profit for end users, such as startups or businesses buying artificial intelligence computing from companies like AWS or Microsoft Azure, which should also make money.
Cloud providers, in particular Microsoft, are actively investing in GPU stocks. Nvidia said that half of its data center revenue comes from large cloud providers, and Microsoft alone is likely to provide about 22% of Nvidia's revenue in the fourth quarter of fiscal 2020. Meanwhile, in the first quarter of fiscal 2020, the company sold about $19 billion worth of graphics processors for data centers.
The introduction of Nvidia B100/B200 processors, promising 2.5 times better performance at a cost of only 25% more expensive, is likely to lead to further investments and create another supply shortage.
According to the analyst, the revenue of OpenAI, which uses Microsoft Azure infrastructure, has increased significantly - from $ 1.6 billion at the end of 2023 to $ 3.4 billion in 2024. This growth highlights OpenAI's dominant position in the market, well ahead of other startups that are still trying to reach the $100 million revenue mark. Nevertheless, investments in artificial intelligence equipment are growing.
According to Kahn, even optimistic forecasts regarding the revenues of large technology companies from artificial intelligence are not justified. If we assume that Google, Microsoft, Apple and Meta earn $10 billion each year from artificial intelligence, while other companies such as Oracle, ByteDance, Alibaba, Tencent, X and Tesla earn $5 billion each, there remains a gap of $500 billion.
