DBS CEO Says AI Infrastructure Risk Repricing Is Underway, With Tougher Profitability Tests Ahead

A broader market repricing of risk in artificial intelligence infrastructure investment is gathering pace, a shift that stands to subject corporate fundraising to tougher profitability scrutiny.

Tan Su Shan, chief executive officer of DBS Group Holdings Ltd., told Bloomberg Television on October 11 that the market is changing how it evaluates AI-related investment. With large sums continuing to flow into the sector, investors are demanding higher returns and more stable funding structures than before.

"The market is already reacting," Tan said. "That is not a bad thing."

Investors want to know whether the huge sums going into AI capital expenditure can generate sufficient long-term return on investment and ultimately deliver value to creditors and shareholders, she said.

The shift could also affect the creditworthiness of AI-related companies. Tan said some companies with weaker financial structures could face credit-rating downgrades, though she described the broader risk reassessment as a healthy development. In her view, it is part of a necessary check on whether the financial burden from large-scale investment is sustainable over the long term.

DBS, Southeast Asia's largest bank, is also expanding its AI-related financing business. Tan said the bank has arranged about S$25 billion, or about $20 billion, of AI-related financing deals this year. Most of those transactions involved DBS serving as lead arranger for syndicated loans, and only a portion of the total consists of loans held directly on the bank's balance sheet.

Demand for funding is concentrated mainly on AI infrastructure buildout. Hardware accounts for most of that, including semiconductors, memory chips and data centers. As capital continues to pour into the AI industry, investment returns and debt-servicing capacity are emerging as key yardsticks for financial markets.