Georges Elhedery, chief executive of HSBC, told CNBC that while artificial intelligence is a major enabler for business, investors must exercise strict discernment when funding the data centers that support it. He warned that not every infrastructure project offers the quality of tenancy agreements or contract pricing necessary for strong investment returns.
Speaking on Tuesday, Elhedery described AI as a "major enabler" that allows banks to re-engineer their activities and boosts employee productivity. He stated that the bank is actively on a journey to adopt the technology to reimagine its operations. However, he drew a sharp line between using AI internally and investing in the external infrastructure boom, advising against an all-in approach to data center financing.
"One has to be selective," Elhedery said, emphasizing the need to scrutinize the quality of tenancy agreements and contract pricing. He noted that some projects lack the level of virtuosity required for strong support, while others may prove more challenging.
HSBC, headquartered in London, operates as both a financier and an operator of data centers. S&P Global ranked it the seventh-largest bank in the world and the biggest lender in Europe last year, with assets valued at close to $3 trillion. The bank's 2025 annual report indicated its intention to become a leading bank for fast-growing transition ecosystems, noting that its customer base includes the AI sector and data centers.
The data center sector has experienced a multibillion-dollar boom as facilities house the servers powering AI and cloud computing. McKinsey estimates global spending on these centers could reach $7 trillion by 2030. To fund this expansion, major technology companies are increasingly turning to private credit and taking on significant debt.
This surge in borrowing is creating tension in financial markets. Mohamed El-Erian, chief economic adviser at Allianz and former CEO of PIMCO, highlighted this pressure at the LSEG and Eurasia Group's GZERO Summit in London earlier this month. He pointed to the competition for financing between governments and AI-related companies as a key source of strain in bond markets.
El-Erian observed that governments are refinancing debt at higher rates while hyperscalers raised $200 billion in bond financing last year. He predicted these companies may raise a trillion dollars in bond financing this year, noting they appear insensitive to yield and spread because they are betting on a transformational technology.
Bond yields from major economies have reached multi-decade highs in 2026. These levels reflect concerns over elevated government deficits, mounting debt, sticky inflation, and more hawkish central bank policies. JPMorgan Chase projected that $4.1 trillion worth of AI-related debt will hit the market before the decade ends as firms continue building capacity.