$180 billion gamble! Can Google's computing Great Wall build an AI empire?

Eastern Time, February 4th, Alphabet (Google’s parent company) released its Q4 2025 (ending December 31) and full-year earnings report, with core financial metrics such as revenue and earnings per share (EPS) significantly surpassing market analyst expectations.

However, behind the impressive performance, the company’s disclosed capital expenditure guidance for 2026 caused a collective shock on Wall Street. Google expects capital expenditures to reach $175-185 billion in 2026, nearly double the total for 2025.

Following the earnings release, Alphabet’s stock experienced a textbook-level “rollercoaster” ride. After hours, the stock price plummeted sharply, dropping by 7.5%, with a market value evaporating approximately $350 billion within minutes; then, as investors delved into the company’s cloud computing profit margins and AI-related order scales, the stock rebounded, with gains exceeding 4% at one point.

In just a few minutes, the total market capitalization of Alphabet fluctuated by about $800 billion. This extreme market reaction not only confirmed investors’ recognition of Google’s current profitability but also starkly exposed Wall Street’s underlying anxiety about the prolonged, costly technological race in the global AI industry.

01 Stellar Performance: Cloud Computing Enters Profit “Harvest” Phase

To understand why Google’s stock could rebound so quickly after a sharp decline, the key lies in its strong commercial monetization capabilities demonstrated this quarter, primarily driven by the deep integration of cloud computing and AI technology.

From a financial perspective, Alphabet achieved a qualitative leap in 2025. Total annual revenue reached $402.8 billion, up 18% year-over-year. On the profit side, Q4 EPS (earnings per share) was $2.82, a 31% increase year-over-year.

Among these, Google Cloud’s performance was particularly outstanding, officially becoming the core engine driving the company’s profit growth. This quarter, cloud revenue hit $17.7 billion, with a further increase in growth rate to 48% year-over-year; more notably, the quality of profitability improved significantly, with operating profit margin soaring from 17.5% in the same period last year to 30.1%.

In the cloud computing industry, a 30% profit margin is a critical industry threshold, indicating that Google Cloud has completely moved past the “burning money to capture market” stage. Leveraging the scale effects and technological advantages of AI large models, it has entered a high-profit return period. Meanwhile, the backlog of Google Cloud orders doubled year-over-year to $240 billion, providing strong certainty for sustained future revenue growth.

On the product side, AI penetration is transforming into impressive data. Currently, Gemini App’s monthly active users (MAU) have exceeded 750 million, and Gemini Enterprise, targeting enterprises, has sold over 8 million paid seats within just four months of launch.

Image source: Google official website

This also marks that this tech giant has successfully transitioned from a traditional ad-driven company to a high-technology barrier provider of computing power and AI services.

Additionally, its strategic positioning within the industry ecosystem was fully validated in this earnings report. As the preferred cloud service provider for Apple’s next-generation foundational models, Google has effectively become a core technological support for the AI transformation of the iOS ecosystem.

Alphabet and Google CEO Sundar Pichai personally confirmed this deep integration during the conference call: “I am pleased to announce that we are collaborating with Apple as its preferred cloud provider and are using Gemini technology to develop the next-generation Apple foundational models.”

This statement implies that the core AI functions in the new iPhone, including model inference and computing power support, will be provided by Google Cloud, allowing Google to directly benefit from Apple’s ecosystem upgrades. This revenue model, based on AI model licensing and computing power leasing, offers higher user stickiness and stronger industry defensibility compared to traditional advertising.

It is precisely these tangible cash flow expectations that, after initial panic over capital expenditure, led investors to once again recognize Google’s ecosystem position in the AI era.

02 Increased Spending of $180 Billion to Strengthen Computing Power Base

Another noteworthy point in the earnings report is Alphabet’s disclosed capital expenditure estimate for 2026, reaching $175-185 billion. This massive investment plan triggered a sharp decline in the stock price after hours.

Investors were evidently panicked; no one knew whether Google’s trillion-dollar-level capital investment was aimed at building a higher industry moat or would fall into a bottomless money-burning pit.

In response to Wall Street’s collective skepticism, Sundar Pichai candidly stated during the conference call: “In fact, the biggest concern keeping us awake at night is the bottleneck in computing power.”

In his view, during the peak of the AI wave, “the risk of under-investment is far greater than over-investment.” This money is not filling a “bottomless pit,” but rather addressing the overflowing customer demand. He revealed that although Google is desperately expanding capacity, it is still in a serious “supply-constrained” state.

This judgment is also the core logic behind Google’s “stockpile first, then defend” wartime mindset, explaining why Google is willing to endure stock price volatility and still push forward with doubled capital expenditure plans.

To better understand this aggressive strategy, it is helpful to compare it with its main competitor, Microsoft. Although both are significantly increasing capital investments, their core approaches to computing power deployment differ sharply. Microsoft’s current industry moat is more built on the deep integration of Azure cloud platform and Office ecosystem, with high capital expenditure mainly supporting core AI computing power for OpenAI, focusing on application deployment and commercialization to generate profits.

In contrast, Google is pursuing a more extreme “vertical integration” route. Its core confidence lies in its self-developed TPU (Tensor Processing Unit). This “self-made chips, self-fighting” capability allows Google to achieve a higher return on investment and bargaining power in its $180 billion “big gamble” compared to competitors heavily reliant on external chip supply.

CFO Anat Ashkenazi pointed out: “Of this nearly $180 billion expenditure, about 60% is allocated to core server chips like computing hardware, and 40% to data centers and power infrastructure for long-term assets.” She emphasized that Google is not blindly throwing money; current investments have already been partially validated through a 30.1% profit margin in cloud services.

Google’s logic is that, in the heavy asset phase of AI, whoever controls the absolute redundancy of power and computing resources will hold the pricing power for the next decade. Sacrificing short-term financial aesthetics for long-term monopoly is also the key reason why Alphabet’s market value rebounded after falling.

03 Google’s Confidence and Challenges

In response to market concerns over the trillion-dollar capital expenditure, Sundar Pichai clearly stated during the earnings call: Google is reshaping and innovating its core business through AI-driven transformation and commercial breakthroughs, validating the reasonableness of this huge investment. The company’s current business performance also supports this confidence.

Google’s core search business AI revolution has completely shattered the prophecy that “AI chatbots will end Google Search.” The quarterly financial data shows that the new “AI Mode” intelligent search feature launched for Search has not cannibalized traditional search traffic; instead, it has enhanced user experience through smarter interactions, leading to higher-quality user engagement.

Data indicates that search queries in AI mode are three times longer than in traditional search, meaning user behavior has shifted from simple keyword searches to complex conversational information research. This behavioral change presents two major commercial opportunities for Google: first, precise ad matching based on deep user needs, significantly increasing ad conversion rates; second, higher traffic monetization efficiency, further enhancing the commercial value of Search.

In this quarter, Google’s Search revenue grew by 17%, enough to prove that AI technology is not disrupting Google’s core business but upgrading it, making its traditional business more profitable.

Moreover, Waymo, Google’s autonomous driving business, has reached a key commercialization turning point, shifting from long-term R&D to a new growth engine. Previously, Waymo was labeled a “money-burning machine” due to continuous R&D investments, but this quarter’s results have surprised the market.

Currently, Waymo has launched its sixth commercial operation market in Miami, with paid autonomous ride services exceeding 400,000 trips weekly, and its commercialization scale continues to expand.

More importantly, AI technology has not only empowered Waymo’s business development but also improved Google’s internal operational efficiency. Using internally developed AI Agent tools, Google now has 50% of its code written and reviewed by AI, significantly reducing R&D costs.

This internal efficiency boost allows Google to support “cost-cutting” (Waymo commercialization and cloud growth) and provides more cash flow for the trillion-dollar capital expenditure.

However, for Alphabet, the operational development in 2026 will face a significant balancing challenge. On one hand, the company needs to accelerate the construction of computing infrastructure to meet the surging demand for AI computing power and services, maintaining its industry leadership; on the other hand, it must leverage high-margin AI software services like Gemini Enterprise subscriptions and revenue sharing from Apple foundational models to rapidly grow revenue and profit, offsetting the financial costs of the trillion-dollar investment.

This earnings report indicates that Google is steadily transforming into a “digital heavy industry empire” focused on building a global computing power foundation. The $400 billion annual revenue confirms its strong monetization capabilities; while the nearly $180 billion capital expenditure guidance signals that the global AI industry has entered a heavy asset phase.

Google’s core challenge in 2026 will be to ensure that the profit growth rate of its AI business keeps pace with the rising costs of building computing infrastructure. This trillion-dollar-scale computing layout will not only determine Google’s future industry positioning but also serve as a key indicator for the development of the global AI industry.