The artificial intelligence industry is expected to generate $6 trillion in annual revenue by 2031, according to a global technology report submitted by consulting firm Bain & Company. This is necessary to continue paying for infrastructure adequate to meet anticipated demand, Report Register.
Achieving these funds will require new, innovative ways to apply artificial intelligence – we are no longer simply talking about making workers more productive. The author of the report stated that the huge demand for artificial intelligence infrastructure has provided new impetus for the hardware industry. HBM’s output is growing rapidly, chip packaging technology continues to develop, and the customized solutions and ASIC markets continue to grow.
Previous reports have suggested that huge spending on artificial intelligence is actually helping the U.S. economy avoid recessionThe construction of data center infrastructure and the development of AI models have become the only truly significant growth vectors. According to Bain & Company, there is an “arms race” among cloud hyperscale companies for artificial intelligence capabilities. Microsoft, Google, Amazon, Meta total capital expenditures✴ Oracle’s sales could reach $780 billion by 2026, nearly five times what they were three years ago.
Photo credit: RONNAKORN TRIRAGANON/unsplash.com
Annual spending on AI infrastructure could reach $1.5 trillion by 2031, roughly in line with Omdia’s forecast. $1.6 trillion By 2030. The $6 trillion figure is based on Bain’s assumption that capital costs will account for about 25% of industry revenue.
It is expected that the application scope of artificial intelligence in traditional fields will bring revenue levels of US$1.2-1.8 trillion. This includes consumer AI services from subscription and advertising revenue, as well as enterprise services for software development, sales, marketing, customer service and IT operations. In other words, it is necessary to find new application areas to create another $4.2 trillion in revenue.
The report identifies four possible categories. Therefore, artificial intelligence models are gradually replacing traditional search engines and providing advertising integration, which will allow them to gain an additional $100-200 billion in revenue. Artificial intelligence in self-driving cars, trucks and drones, as well as industrial automation systems, will help create new products and services, generating an estimated $400 billion in revenue. The field of “physical” artificial intelligence with analog, digital twins, and robotics could generate up to $900 billion in revenue. Up to $1.5 trillion in funding could be available in just three categories.
The remaining funding must come from “new products and applications that don’t yet exist.” This category includes the use of artificial intelligence to create new medicines, mental health support services (a billion-dollar unmet need), and new solutions in materials science (particularly batteries and semiconductors). In addition, research in industries ranging from “neuroscience” to fusion energy is expected to accelerate.
Analysts stress that modern discussions have largely focused on improving worker productivity, but the new economy of artificial intelligence infrastructure requires huge sums of money and cannot provide simple performance improvements. Experts say the industry needs a “wave of innovation” greater than the economic changes brought about by the emergence of mobile technology and cloud computing.
A Bain & Company report released a year ago predicted that by 2030, the industry should offer $2 trillion revenue to drive infrastructure development. This figure has tripled, indirectly showing the increase in investment in artificial intelligence in the past 12 months. The purported $6 trillion figure may not be achieved as experts have repeatedly said that the implementation of artificial intelligence has not yet delivered the expected financial returns.
Additionally, not all AI infrastructure projects will be implemented. In fact, the U.S. is building only half Of the production capacity planned to be built in 2026 and the projects planned to be put into operation in 2028, 80% have not yet started construction. In a separate report, the company noted that limitations in chip manufacturing will limit data center deployment in the coming years.
source:
