Nikkei 225 Falls 2.73% as Alphabet AI Costs Shake Japan Tech Stocks

Key Takeaways

• The Nikkei 225 closed 2.73% lower at 64,611.15 on July 24 as Japanese semiconductor and AI-related shares faced renewed selling.
• Alphabet’s quarterly capital expenditure reached $44.9 billion, pushing free cash flow to negative $5.9 billion despite rapid Google Cloud growth.
• Advantest, Tokyo Electron, SoftBank Group and Kioxia led the technology decline, while railway, shipping and pharmaceutical shares provided limited support.

Alphabet’s AI Spending Shock Reaches the Nikkei 225

Japan’s Nikkei 225 fell sharply on Friday as Alphabet’s expanding artificial intelligence budget renewed concerns about the cost of sustaining the global AI boom.

The index closed at 64,611.15, down 2.73%, after falling as low as 64,201.03 during the session. It has now lost more than 7% in July and remains under pressure after entering correction territory earlier in the month. Official Nikkei index data showed that the benchmark opened at 65,584.25 but failed to hold above 65,000.

The broader Topix also declined, although its losses were smaller than those of the Nikkei. This difference reflected concentrated selling in the high-priced technology companies that have a disproportionate influence on the price-weighted Nikkei 225.

Sentiment weakened after Alphabet shares dropped around 7.2% to $317.69 in New York. The decline followed a wider retreat across US technology stocks, with the Nasdaq Composite falling more than 2%.

The market reaction suggested that investors are no longer evaluating AI companies solely on revenue growth. Capital requirements, free cash flow and the time needed to generate returns from new infrastructure are becoming increasingly important.

Why Strong Google Cloud Growth Failed to Reassure Investors

Alphabet’s second-quarter results provided clear evidence that demand for AI services remains strong.

Total revenue increased 24% year on year to $119.8 billion, while Google Cloud revenue surged 82% to $24.8 billion. Google Services revenue rose 15% to $94.5 billion, and the company’s operating margin expanded to 34%.

However, investors focused on how much Alphabet had to spend to support that growth.

Quarterly capital expenditure climbed to $44.9 billion, exceeding the $39.1 billion generated through operating activities. This left Alphabet with negative free cash flow of approximately $5.9 billion for the quarter.

Negative free cash flow does not mean Alphabet was unprofitable. Instead, it indicates that infrastructure investment exceeded the cash generated from operations during the period. The company still reported a 30% increase in operating income. Alphabet’s official Q2 results nevertheless showed how quickly AI-related expenditure is consuming available cash.

Alphabet also raised its expected 2026 capital expenditure to between $195 billion and $205 billion, compared with its previous forecast of $180 billion to $190 billion.

Management said additional computing capacity was necessary because customer demand continued to exceed supply. For the market, however, the central question is how quickly this new capacity can produce enough revenue and profit to justify the investment.

Japan’s AI Hardware Leaders Bear the Brunt

Alphabet’s share-price decline quickly spread to Japanese companies connected to the AI hardware supply chain.

During morning trading, semiconductor-testing equipment manufacturer Advantest fell 6.33%, while Tokyo Electron declined 5.43%. SoftBank Group dropped 7.42%, and memory producer Kioxia lost 4.4%. Advantest’s decline later extended to more than 7%.

These companies occupy different parts of the AI ecosystem. Advantest supplies semiconductor testing equipment, Tokyo Electron provides chip-manufacturing machinery, and Kioxia produces memory used across data-intensive computing systems. SoftBank Group also carries substantial exposure to technology and AI investment themes.

Their declines did not indicate that demand for AI chips had suddenly disappeared. Alphabet’s accelerating cloud growth and its statement that computing demand continues to exceed capacity suggest the opposite.

Instead, the sell-off reflected growing sensitivity to the economics of the AI investment cycle. If technology companies must continue increasing expenditure faster than they generate cash, investors may apply lower valuation multiples until the eventual return on that spending becomes clearer.

The structure of the Nikkei 225 amplified the effect. Because the index is price-weighted, large movements in high-priced technology shares can move the benchmark more sharply than changes in the broader Japanese market.

Domestic Shares Provide Limited Protection

The session was not defined by indiscriminate selling across every part of the Japanese market.

Railway stocks advanced as investors moved towards companies with more domestic revenue and less direct exposure to AI valuations. Central Japan Railway gained 1.17% during morning trading, while East Japan Railway added 0.6%.

Shipping companies also performed relatively well. Kawasaki Kisen and Mitsui OSK Lines rose 0.61% and 0.88%, respectively. Pharmaceutical company Otsuka Holdings gained approximately 1.6%, making it one of the strongest performers within the Nikkei.

This divergence pointed to a rotation rather than a complete withdrawal from Japanese equities. Investors appeared to be reducing exposure to expensive technology shares while favouring businesses with more predictable cash flows.

However, higher energy prices created an additional risk for the broader market. Brent crude moved above $100 per barrel as Middle East tensions intensified, increasing concerns about imported inflation and higher operating costs for Japanese companies.

What Could Move the Nikkei 225 Next?

Japanese corporate earnings will be important in determining whether the Nikkei 225 can stabilise.

Strong guidance from domestic semiconductor, industrial and technology companies could help investors separate Japan’s underlying earnings outlook from concerns surrounding US AI expenditure. Weak forecasts would reinforce fears that valuations have moved ahead of near-term profit growth.

Upcoming results from other major US technology companies will also be closely watched. Investors will compare cloud revenue, AI-related sales and capital spending to determine whether Alphabet’s negative free cash flow was company-specific or part of a broader industry pattern.

The most important measure may be the relationship between AI expenditure and cash generation. Rapid revenue growth can support high infrastructure spending, but persistent negative free cash flow could increase pressure on margins, financing requirements and shareholder returns.

Oil prices and the Japanese yen remain additional factors. Expensive energy and a weak currency can raise import costs, complicate the inflation outlook and reduce the benefit that exporters usually receive from yen depreciation.

Alphabet Exposes the AI Boom’s Capital-Intensity Risk

Alphabet’s results did not show that the AI boom lacks customers. Google Cloud’s 82% revenue growth and continued shortage of computing capacity demonstrated strong underlying demand.

The more expensive problem is that meeting this demand requires enormous and recurring investment in servers, data centres, networking equipment and energy infrastructure. Revenue may be expanding rapidly, but the required spending is expanding just as quickly.

For the Nikkei 225, that uncertainty is especially important because its recent performance has become heavily dependent on a small group of AI and semiconductor companies.

The July 24 sell-off therefore represented more than a reaction to one company’s earnings. It reflected a wider reassessment of how long the AI investment cycle may take to generate sustainable cash returns—and how much investors are willing to pay while they wait.