The India Decade

DBS CEO Tan Su Shan says market is repricing AI financing risks

Tan Su Shan says investors are demanding stricter return on investment and sturdier debt structures across massive technology capital expenditure.

By The India Decade

server racks inside a modern commercial data centre (file image)
server racks inside a modern commercial data centre (file image) · “Data Centres” by GDS Infographics (BY 2.0) via Flickr

Financial markets have begun repricing the credit risks attached to artificial intelligence investments, demanding higher returns and far stricter underwriting before funding massive technology projects, according to DBS Group Holdings Chief Executive Tan Su Shan.

Speaking in an interview with Bloomberg Television, the head of Southeast Asia's largest lender described the shift towards tighter financing discipline as a necessary correction for an industry absorbing unprecedented sums of capital.

"The market has spoken," Tan said. "It is not a bad thing because now, people are asking for higher returns, and people are also looking at structures. They are making sure that whatever money is now being hoovered into the whole AI capex ecosystem will lead to some kind of long-term ROI that can then be repaid back to debt or bond or equity holders."

What the repricing means for borrowers

The reassessment comes as global technology firms commit hundreds of billions of dollars to computational infrastructure against a backdrop of elevated bond yields and volatile interest rates. Tan warned that borrowers with weaker balance sheets could face credit-rating downgrades as financiers examine whether cash flows can genuinely service their debt commitments over the long haul.

Yet she stressed that tighter scrutiny should be welcomed rather than feared. By distinguishing viable projects with genuine paths to cash generation from speculative ventures, the correction helps insulate broader capital markets from systemic stress.

Across Asia, the scale of capital required remains immense. PricewaterhouseCoopers estimates that total spending on data centres across the region could reach US$8.2 trillion by 2050, requiring deep, stable pools of domestic and foreign institutional funding.

DBS arranges S$25 billion in AI debt

DBS has been active across the sector, originating approximately S$25 billion (around $19.5 billion) in AI-linked transactions so far this year. However, the bank has retained only a small fraction of that exposure on its balance sheet.

Instead, the Singapore-headquartered group has acted predominantly as lead manager on syndicated loan facilities, packaging the debt and distributing portions across international institutions. Tan said the financing demand has concentrated almost entirely on physical hardware, ranging from semiconductor fabrication and memory chips to high-capacity data centre developments.

Beyond external financing, DBS is using AI tools internally to automate workflows, accelerate account openings, and expand fee revenues after Tan took the helm in early 2025. While she acknowledged severe interest rate swings and mounting geopolitical tensions in eastern Europe and the Middle East, she noted that wealth creation across Asia continues to support stable long-term capital flows.

Key numbers

WhatFigureSource
DBS AI deal origination this yearS$25 billionTan Su Shan
Projected Asia data centre spending by 2050US$8.2 trillionPricewaterhouseCoopers LLP

In this story

NameRole in the story
Tan Su ShanChief Executive Officer of DBS Group Holdings who commented on the repricing of AI debt
DBS Group HoldingsSoutheast Asia's largest bank by assets, which has originated S$25 billion in AI-related deals this year

Topics

The day’s reporting, once each morning. No advertising.

More on this story