AI Boom Or Bubble? Why Massive Tech Spending May Not Pay Off Soon

Global tech leaders such as Microsoft, Amazon, Alphabet, and Meta are pouring unprecedented capital into artificial intelligence infrastructure, according to a new report by Jefferies.

The report, titled "The mother of all capex cycles", estimates that AI-related capital expenditure could hit USD 700 billion this year and USD 800 billion next year. This reflects the scale of a global race to build computing power, data centres, and advanced AI models.

"This number represents about 2% of US GDP and about 20% of US non-residential fixed investment," the report noted, adding that it is also "equivalent to nearly 30% of total non-financial pre-tax profits of all US companies."

Why are concerns rising despite massive investments?

Even as spending accelerates, concerns are mounting about whether these investments will generate sustainable returns. The report highlights growing pressure on company cash flows.

"Capex as a percentage of operating cash flow has risen from 41% in 2023 to a projected 92% in 2026," the report said, underlining how AI investments are increasingly consuming company resources.

This sharp rise suggests that companies are prioritising long-term AI leadership over short-term profitability, a strategy that carries financial risks.

Is AI proving difficult to monetise?

A major concern flagged by Jefferies is the lack of clear revenue models in AI. While demand for AI tools is rising, turning that demand into consistent profits remains challenging.

"Rising investment required to maintain leadership... will likely mean 'sustainable profitability is far away for pure model players'," the report said.

This indicates that companies focused primarily on building AI models—rather than diversified business ecosystems—may struggle the most.

Could AI become a capital-heavy industry?

The report challenges the popular belief that AI will follow the high-margin, winner-takes-all dynamics of the internet era.

"AI will turn out to be much more like the capex-intensive airline industry than the winner-takes-all network effect of the Internet economy," the report observed.

If this proves true, AI could become an industry where heavy investment is required just to stay competitive, limiting profitability.

Are tech companies slowing down their spending?

Despite investor concerns, major firms are not pulling back. Instead, they are doubling down on AI investments.

"Microsoft said... it expects to spend USD 190 billion in capex this calendar year," the report noted, while Alphabet and Meta have also raised their spending outlook.

This suggests that companies fear falling behind in the AI race more than they fear short-term financial strain.

Are there cracks in the AI growth story?

The report also points to early signs of stress in the AI ecosystem, including missed targets and intensifying competition.

A recent media report cited by Jefferies noted that OpenAI "missed an internal goal of reaching 1bn weekly active users" and has "missed multiple monthly revenue targets earlier this year."

At the same time, competition is heating up. Gemini has gained traction, while ChatGPT has seen its market share decline.

"Gemini's web traffic share... has increased from 6% to 25.5%... while ChatGPT's market share has declined from 77.4% to 56.7%," the report said.

Who is benefiting from the AI boom?

While AI developers face uncertainty, suppliers of critical components—especially memory chips—are emerging as clear winners due to strong demand and limited supply.

This highlights how infrastructure providers may profit more reliably than AI application developers in the current phase.

What is the biggest risk going forward?

Jefferies warns that the biggest threat is a potential reassessment of the entire spending cycle.

"The risk... is a sudden realisation by hyperscalers, or investors, that they have over invested," the report said.

Such a shift could trigger spending cuts, impacting the broader tech ecosystem.

(With inputs from ANI)