Sound the AIarm: Tech stocks’ whiplash could signal risks for Singapore’s real economy

Investors pay a premium for growth, but market swings reveal unease about incentives driving the boom

[SINGAPORE] When Monetary Authority of Singapore (MAS) managing director Chia Der Jiun delivered his speech at the release of the central bank’s annual report for FY2026 on Tuesday (), his remarks might have been drafted as a routine macroeconomic warning.

Instead, they served as live commentary for a storm breaking in real time.

Among other things, Chia flagged the artificial intelligence investment boom as a massive, looming uncertainty.

“While near-term investments are supported by committed orders and strong hyperscaler cash flows, there is greater uncertainty around the sustainability of these investments in the medium term,” he said.

A day earlier on Jul 27, Fitch Ratings sounded the same alarm.

In its latest outlook, the agency classified an AI market correction as a major short-term global credit risk, putting it on the same level as persistent geopolitical tensions in the Middle East.

The scale of the exposure Fitch described is immense. US IT capital expenditure grew 18 per cent year on year in the first quarter, adding 1.4 percentage points directly to US GDP growth and helping to fuel a 26 per cent surge in corporate bond issuance in the first half of the year.

Amazon, Alphabet, Nvidia, Meta, Oracle and SpaceX alone issued US$182 billion of investment-grade bonds.

Rising tech valuations have also propped up US household spending through the wealth effect, even as Fitch cut its 2026 US growth forecast to 1.9 per cent and lifted its year-end inflation estimate to 3.7 per cent on the back of the oil shock from the Middle East conflict.

Hyperscalers have announced data centre spending running into the hundreds of billions of dollars.

This aggressive spending spree boosted Asian economic performance, driving high export growth on the back of the semiconductor boom.

But MAS’ Chia sketched out a clear risk: What happens if the market increases the cost of financing because it stops believing this technological race can be won at a reasonable cost?

Vulnerability exposed

Just as Chia delivered his warning, chipmaker stocks entered the throes of a bruising three-day sell-off that erased more than US$1 trillion in market value.

This came after reports emerged that a Shanghai-based, state-backed manufacturer has begun mass-producing immersion deep ultraviolet lithography machines.

Together with negative cash-flow updates from Alphabet and Tesla the week before, this triggered a fresh bout of panic selling. Wall Street suddenly wanted proof of commercial revenue, and chip valuations paid the price.

Then came Jul 30, and the market simply changed its mind.

Microsoft released a stellar forecast that seemingly washed away fears of excessive infrastructure spending. The stock rocketed more than 15 per cent – its best single day since 2008 – adding close to US$450 billion to its market value in one session.

This was the largest one-day dollar gain any company has recorded, eclipsing Nvidia’s own record set last year.

Amazon delivered a similar jolt that same evening: AWS’ revenue grew 37 per cent, its fastest pace since 2021, and the stock jumped more than 9 per cent in after-hours trade even as the group raised its 2026 capital spending forecast to US$220 billion.

The message from both: AI spending is starting to show up as revenue, not just cost.

The euphoria bled into Jul 31’s Asian session, reviving confidence in the AI trade.

South Korea’s Kospi – a bellwether for AI sentiment – surged by a record 18 per cent, as SK Hynix and Samsung Electronics pulled out of the slump with gains of close to 30 per cent apiece.

In Singapore, the local semiconductor trio of AEM Holdings : AWX +3.62%, UMS Integration : 558 +0.42% and Frencken Group : E28 +0.78% rebounded too.

But this whiplash is exactly the vulnerability MAS’ Chia and Fitch were warning about. The market is trading on faith, and that faith is proving deeply erratic.

Stakes are high

The volatility goes beyond stock price swings; don’t forget the wave of institutional money chasing the AI narrative. These investors are paying a hefty premium for growth, but last week’s swings reveal a growing unease about the incentives driving the boom.

At the same time, Singapore’s export engine is wired directly into the underlying cycle.

June’s non-oil domestic exports (NODX) rose 20.7 per cent year on year, with electronics exports alone surging 105 per cent, supported by robust AI-related demand.

Growth in this segment was mainly driven by integrated circuits, which expanded 115.4 per cent, reflecting the global demand for memory chips.

Electronics NODX growth was also driven by a 170.9 per cent rise in disk media products, and a 95.8 per cent expansion in the exports of personal computers.

But as Chia outlined in his address, two distinct paths lie ahead.

A continued boom will spill into higher demand, complicating life for central banks as they manage inflation. A sharp retreat, on the other hand, will hit business investment and semiconductor demand directly.

Either way, the stakes are high.

This aligns with Fitch’s own warning: A larger, protracted correction will risk wider contagion, exposing financial stability risks where credit markets have leaned on shaky business models.

For now, analysts are calling last week’s wild swings a mid-cycle reset rather than the start of something worse.

Indeed, the hard data has not yet shown the demand destruction that would prove the bears right, and the next export print will show whether AI demand is still turning into actual orders.

Until then, investors in Singapore are stuck watching a deeply bipolar market.

These three listed semiconductor-related names are only the most visible tip of a much deeper exposure; the wider technology ecosystem here is staring down the same volatility.

The trade data insists the AI party is in full swing, while the stock market is having a full-blown mood swing.

But the warnings from Chia and Fitch boil down to a simple reality check: You cannot run a physical supply chain on blind faith forever. When the market decides the AI race costs too much to fund, the orders for the parts that build it could stop too.

source: The Business Times https://www.businesstimes.com.sg/companies-markets/sound-aiarm-tech-stocks-whiplash-could-signal-risks-singapores-real-economy