Asia’s 8% Weekly Drop Shows the AI Stock Selloff Is Spreading
1. Takeaway
The AI stock selloff intensified Friday, with Asia-Pacific shares outside Japan heading for an 8% weekly decline, their steepest weekly drop since March 2020, according to Reuters’ Friday market report. U.S. Nasdaq 100 futures were also lower in the premarket, while investors questioned whether the enormous sums being spent on AI infrastructure will produce returns quickly enough to justify current valuations. The immediate conflict for investors is that strong company results and major new AI listings have not been enough to restore confidence in the sector.
2. What happened?
Technology shares came under pressure worldwide as investors reassessed AI investment and future profits. The selloff continued even after Anthropic’s high-profile stock-market debut on Thursday. Anthropic priced its shares at $83, opened at $111.50 and closed its first day at $97.17, up 17% from the offer price, but the gains faded during the session and the stock fell further in Friday premarket trading. The debut had initially raised hopes about investor demand for AI companies, but it did not reverse concerns about the wider technology trade.
Reuters reported that the MSCI global equity index was heading for a seventh straight daily decline. Asian technology markets were among the weakest areas. Japan’s Nikkei fell about 5% on Friday. Taiwan’s market faced a weekly decline of roughly 17%. Reuters called it Taiwan’s steepest weekly fall since 1990. The same report put the weekly loss for Asia-Pacific shares outside Japan at 8%, the largest since March 2020. These are regional market readings, not U.S. index returns.
The concern is not simply that technology shares have become expensive. Investors are also debating whether major technology companies can convert their large capital spending into enough revenue, cash flow and productivity gains. Data centers, advanced chips, power systems and networking equipment require large upfront commitments. The benefits may take years to appear, while depreciation, energy costs and financing expenses begin sooner.
Another source of pressure was a report that U.S. authorities may scrutinize AMD’s China ties. AMD shares fell 8% on Thursday after Hunterbrook Media reported that IEIT Systems, a Chinese server maker blacklisted by the U.S. Department of Commerce in March 2025, had incorporated newer versions of AMD’s firmware. Reuters could not independently verify the report, and AMD denied the claims. The reporting available before the U.S. open did not establish the details or potential consequences, so treat the issue as a reported risk rather than a confirmed new restriction.
3. The key numbers
- 8%: Weekly decline for Asia-Pacific shares outside Japan, according to Reuters’ Friday report.
- About 5%: Friday decline in Japan’s Nikkei, as reported by Reuters.
- About 17%: Taiwan market’s weekly decline, which Reuters described as its steepest since 1990.
- Seven sessions: The global equity index was on course for its seventh consecutive daily decline, according to Reuters.
- 1.5%: Nasdaq 100 futures decline cited in Reuters’ Friday market coverage. This is a premarket futures reading, not the Nasdaq 100 cash index’s Friday performance.
4. Why does it matter?
The global nature of the selloff matters because AI investment extends beyond a small group of U.S. software companies. Semiconductor manufacturers, chip-equipment suppliers and memory producers share the same spending cycle. So do data-center operators, cloud providers and electricity infrastructure businesses. When investors question the expected return on AI infrastructure, the valuation pressure can spread across that entire chain.
A decline in chip shares can also affect companies that are not direct AI businesses. Broad technology indexes contain firms whose revenue depends on advertising, online commerce, enterprise software, smartphones and cloud services. If investors reduce exposure to technology as a group, shares can fall even when a particular company’s latest results are not the source of concern.
The other important issue is the difference between earnings performance and expectations. A company can report strong revenue growth and still see its share price decline if investors had expected even faster growth. In a high-valuation market, the question is not only whether a business is growing. It is whether the growth is sufficient to support the price investors are already paying.
That distinction may become more important as companies prepare to discuss capital spending for the coming year. Investors will want to know whether AI-related spending is generating measurable returns, whether customers are paying for the new services, and whether demand is broad enough to support the infrastructure being built.
5. What the market is debating
The central disagreement is between investors who see AI infrastructure as a long-term economic transformation and investors who believe expectations have moved too far ahead of demonstrated profits.
The optimistic case is that demand for computing capacity will continue to expand. Companies are integrating AI into search, productivity software, customer service, cybersecurity, industrial operations and data analysis. If customers pay for these services and businesses use them to reduce costs or increase output, infrastructure spending could support a long period of revenue growth.
The cautious case is that spending may grow faster than monetization. Building data centers and buying advanced chips does not automatically create profitable products. Companies must also pay for electricity, cooling, networking, maintenance and the employees needed to operate complex systems. If customers delay adoption or resist higher prices, the expected returns could take longer to arrive.
There is also a question about the distribution of those returns. A large technology company may capture much of the economic value from AI services, while suppliers face intense competition and pressure on margins. Investors therefore cannot assume that every business connected to AI will benefit equally.
6. What it means for investors, employees and customers
For investors, the latest market action is a reminder that exposure to a popular theme can carry risks even when the underlying technology continues to advance. Share prices reflect expectations about future cash flows, not just the importance of a product or the amount of money being invested in it. A company can remain strategically important while its stock valuation falls.
For employees, a prolonged period of weaker technology valuations could influence hiring plans, compensation packages and the pace of expansion. Private companies may find fundraising more difficult if public-market investors become less willing to pay high prices for growth. Public companies may also face pressure to demonstrate tighter spending discipline. These are possible business responses, not confirmed layoffs or hiring decisions.
For customers, the immediate effect may be less visible. A market repricing does not automatically mean AI products will become cheaper or less capable. However, if providers slow infrastructure investment or focus on profitability, customers could see changes in product pricing, service availability or the speed at which new features are introduced.
For suppliers, the distinction between committed orders and speculative demand is especially important. Equipment makers and data-center contractors need to know whether projects are funded and moving forward, rather than relying only on broad statements about future AI demand.
7. The earnings and economic calendar
Friday’s earnings calendar includes several companies scheduled to report, including BJ’s Wholesale Club, Moog, Frontline and IES Holdings. Reporting times and the complete company list vary across calendar providers, so investors should confirm the release time directly with each company before relying on a calendar entry.
The BLS has October state employment and unemployment data and annual 2025 total factor productivity data scheduled for 10:00 a.m. ET Friday. These releases are due later in the morning and should not be described as published results at the 7:00 a.m. ET cutoff.
The Federal Reserve calendar also lists a speech by Governor Christopher Waller at 8:25 a.m. ET and a speech by Governor Lisa Cook at 12:15 p.m. ET. Investors will listen for comments about inflation, economic activity and the policy outlook, but scheduled remarks do not establish that the Fed has made a new rate decision.
8. What to watch next
The first test is whether U.S. technology shares follow the weakness seen in Asian markets when regular trading begins. Futures can change before the opening bell, so the premarket reading should not be treated as a forecast of the final session.
The second test is whether the decline remains concentrated in high-growth technology shares or spreads into banks, industrial companies, consumer businesses and smaller stocks. A broader retreat would suggest that investors are reducing overall risk rather than simply rotating away from AI-related names.
The third test is company guidance. Investors should watch for comments about data-center spending, chip demand, cloud revenue, customer adoption and expected returns on AI investment. Management teams may face more direct questions about the time required to turn spending into profits.
The fourth test is whether market volatility changes financing conditions. If investors demand higher returns for holding risky assets, companies with large investment programs may face greater pressure to justify their funding plans. That pressure could affect private firms as well as listed companies.
9. Bottom line
The global AI stock selloff is becoming a test of earnings expectations, not simply a reaction to one company’s results. The weakness in Asian technology shares and U.S. futures suggests that investors are questioning the economics of the entire AI investment cycle. The technology may continue to develop rapidly, but the market is demanding clearer evidence that spending will translate into durable profits.
For U.S. investors, the next phase will depend on whether companies can demonstrate measurable customer demand and returns on infrastructure spending. Friday’s economic releases and Federal Reserve remarks may influence the wider market backdrop, but company guidance and evidence of monetization remain central to the AI valuation debate.
10. Sources and further reading
- Reuters, November 20, 2026 — Stocks slide as AI selloff deepens, Asia set for worst week since 2020.
- U.S. Bureau of Labor Statistics — November 2026 release schedule.
- Federal Reserve — November 2026 calendar.
This article is for general informational and educational purposes only and is not financial advice or a personal investment recommendation. Market data can change quickly — check current figures before acting on anything you read here.
Read also: Global Market Stress Tests AI Valuations, Borrowing Costs and Crypto Demand




