The race to invest in artificial intelligence (AI) is heating up, with the largest US hyperscalers on track to spend a staggering $800 billion this year alone. According to consensus estimates, this number will continue to grow, reaching $1.1 trillion in 2027 and an astonishing $1.4 trillion by 2028. However, the growth of AI investment is not without its challenges. As capex continues to outstrip operating cash flow, more of the buildout will need to be funded through debt or equity.

The biggest constraint on AI investment, according to Goldman Sachs (GS), is the market capacity for investment-grade debt. GS estimates that hyperscalers will need around $300 billion in annual AI revenue to break even on their 2026-27 investment. While cloud revenues are accelerating, with announced backlogs exceeding $1.5 trillion, the real challenge lies in convincing users to spend around $1 trillion on AI applications each year. This is a significant portion of the global software market, which currently stands at around $1.5 trillion.

In order to make solid returns and for AI apps to generate strong margins, GS estimates that users will need to spend around $1 trillion on AI applications each year. This is a tall order, but one that the industry must work together to achieve. The future of technology hangs in the balance, and it will require significant investment and innovation to unlock its full potential.

As the race to invest in AI continues, it will be interesting to see how the industry addresses these challenges. Will hyperscalers be able to find the necessary funding to support their ambitious plans? Can users be convinced to spend more on AI applications? Only time will tell, but one thing is certain: the future of technology hangs in the balance, and it will require significant investment and innovation to unlock its full potential.