A warning has emerged that a race among big tech companies to invest in AI infrastructure is threatening credit quality by worsening free cash flow and increasing pressure on balance sheets.

On July 24 local time, CNBC reported that Moody's assessed that the financial structures of six hyperscalers - Microsoft, Amazon, Alphabet, Meta, Oracle and CoreWeave - are being strained by unprecedented AI spending.

Moody's forecast that their capital spending will reach $785 billion in 2026 and rise to about $1 trillion in 2027. It cited a sharp jump in demand for investment in physical infrastructure such as data centres, servers and chips as generative AI spreads.

In the past, these companies maintained high profitability with relatively little capital, based on software, intellectual property and scalable cloud services. But as expanding AI infrastructure requires large upfront investment, they are rapidly shifting from asset-light to asset-heavy structures, CNBC reported.

Funding methods are also changing. The six companies' direct debt was tallied at about $460 billion. Even cash-rich companies such as Alphabet and Microsoft are relying more on borrowing, stock sales and off-balance-sheet financing. Alphabet last month announced a stock sale plan worth $85 billion.

Moody's also pointed to risks from a circular structure within the AI investment ecosystem. Some hyperscalers' order backlogs of billions of dollars come from strategic deals with private AI research labs such as OpenAI and Anthropic. The companies invest in the labs, and the labs in turn buy the same companies' cloud services on a large scale.

Even so, demand for AI computing remains strong and cloud businesses continue to grow. Moody's forecast that investors will look more closely at these companies' ability to secure returns on investment going forward.