Global markets extended a technology-led retreat Friday as investors continued selling semiconductor companies and other major beneficiaries of the artificial-intelligence boom. Stocks tumbled across several Asian markets before U.S. trading opened lower, reinforcing concerns that a reassessment of AI valuations was spreading through the international financial system.

In Friday’s Asian session, Taiwan’s benchmark dropped 6.5%, Tokyo fell 4% and Shanghai lost 3%, according to the Associated Press. Taiwan Semiconductor Manufacturing Co., the world’s largest contract chipmaker and a critical supplier to U.S. technology companies, declined 7.3%. European indexes moved more modestly, reflecting the region’s lower concentration of large AI and semiconductor stocks.

On Wall Street, the S&P 500 was down 0.5% around midday Friday after falling as much as 1.4% earlier. The Dow Jones Industrial Average was off 149 points, or 0.3%, while the Nasdaq composite declined 0.7%. The indexes recovered part of their morning losses, but the trading remained volatile and left the S&P 500 on course for its first losing week in three.

The moves followed a sharp decline Thursday, when the S&P 500 lost 0.5% to close at 7,533.77. The Dow fell 105.67 points, or 0.2%, to 52,552.97, while the technology-heavy Nasdaq dropped 1.5% to 25,881.95. Nvidia declined 2.4% and was the largest drag on the S&P 500 because of its enormous market value and corresponding influence on the capitalization-weighted index.

Other companies tied to AI computing and data storage suffered steeper losses Thursday. Micron Technology fell 5.6%, Sandisk dropped 12.6% and Western Digital sank 9.2%. Those declines came after exceptional advances earlier in the year, leaving investors debating whether the pullback represented ordinary profit-taking or a more durable change in expectations for AI hardware demand.

The market’s internal performance was less negative than the headline indexes suggested. Nearly three out of every four companies in the S&P 500 rose Thursday, according to the AP. Abbott gained 10.7% after reporting stronger-than-expected profit and raising its full-year earnings forecast, while J.B. Hunt Transport Services advanced 8% following better quarterly results. Those gains were not enough to offset losses among the index’s largest technology companies.

Financial market screens display falling technology and semiconductor stocks during a global AI-led selloff.

That divergence illustrates the concentration risk created by the rise of AI stocks. When a company such as Nvidia falls, the effect on the S&P 500 can be greater than the combined movement of numerous smaller companies. Investors tracking broad indexes therefore remain heavily exposed to changes in sentiment toward AI processors, cloud-computing infrastructure and the capital-spending plans of major technology platforms.

The retreat persisted even after TSMC reported strong second-quarter results. The company said revenue reached $40.20 billion, up 33.7% from a year earlier, while net income increased 77.4%. Advanced manufacturing technologies accounted for 77% of wafer revenue. TSMC shares rose 1.2% in Taiwan immediately after the report Thursday, but its U.S.-traded shares fell 2.3%, before the sharper decline in Taipei on Friday.

The response suggested that strong current demand was no longer sufficient to remove longer-term concerns. Investors are examining whether spending by cloud providers and AI developers can continue growing at its recent pace, whether data-center customers will generate adequate returns and whether cheaper or more efficient models could reduce the amount of computing equipment required for some tasks.

Those concerns intensified after the release of Kimi K3, an open-source AI model from Chinese startup Moonshot. The development revived memories of the market reaction to DeepSeek in early 2025, when investors questioned whether capable models developed at lower cost could weaken demand for the most expensive processors and infrastructure. The long-term effect of Kimi K3 on chip consumption remains uncertain, but its introduction added another reason for investors to reassess aggressive growth assumptions.

South Korea has been one of the clearest examples of AI-related volatility. The Kospi fell 6.4% Thursday as Samsung Electronics and memory-chip producer SK Hynix declined. The index had risen 6.2% the previous day and has also recorded several unusually large drops in recent weeks. South Korea’s market was closed Friday for a holiday, temporarily pausing the swings.

Korean shares also faced pressure from monetary policy. The Bank of Korea raised its base rate by 25 basis points Thursday, from 2.50% to 2.75%, its first increase since 2023. The central bank cited stronger growth, inflation that was expected to remain above target and continuing financial-stability risks. It also said stock prices had experienced a sizable correction amid concerns about AI investment and foreign selling.

Financial market screens display falling technology and semiconductor stocks during a global AI-led selloff.

The rate decision matters beyond South Korea because investors are again considering whether the Federal Reserve and other central banks could tighten policy in response to higher energy costs. Rising interest rates reduce the present value of expected future earnings, a dynamic that can weigh especially heavily on fast-growing technology companies whose valuations depend on profits anticipated many years ahead.

Oil added another source of pressure Friday. Brent crude rose 3.9% to $87.48 a barrel, up from roughly $76 a week earlier, as conflict involving Iran increased concerns about shipments through the Strait of Hormuz. Expensive oil can raise transportation, manufacturing and consumer costs, complicating central banks’ efforts to control inflation.

Treasury yields eased somewhat Friday, with the 10-year yield falling to 4.54% from 4.57% late Thursday. Even so, borrowing costs remain elevated, affecting corporate financing, mortgages and investment decisions. A sustained combination of high yields and declining technology valuations could make it more expensive for AI startups, data-center developers and equipment suppliers to raise capital.

U.S. economic data offered a mixed backdrop. June retail and food-service sales rose 0.2% from May to $768.6 billion, according to the Census Bureau, while initial unemployment claims fell to 208,000 in the week ending July 11. The figures indicated that consumer spending and the labor market remained resilient, but they did not resolve questions about inflation or the future path of interest rates.

The immediate market debate is no longer simply whether AI demand is growing. Investors are asking whether that growth can meet expectations already embedded in stock prices and whether companies can convert record infrastructure spending into durable earnings. The latest global selloff shows that even strong operating results may provide limited protection when confidence in an entire investment theme begins to weaken.

Leave a Reply

Trending