This article is crafted based on publicly accessible information and is intended solely for the purpose of informational exchange, not as any form of investment advice.

On July 22, 2026, after the market closed, capital markets imposed a paradoxical narrative constraint on Google.

The Gemini app has amassed 950 million monthly active users, marking a 46% surge from the 650 million recorded in October of the previous year. Yet, in the same quarter, Alphabet witnessed its first negative free cash flow since going public in 2004, with an outflow of $5.9 billion. The upper limit of its capital expenditure guidance was revised upwards to $205 billion, and the flagship model Gemini 3.5 Pro, initially slated for a June release, remains ensnared in partner testing.

The market swiftly provided a straightforward explanation: Google is indulging in reckless spending, yet the returns are lagging, and even its flagship model is trailing behind. The stock price took a nosedive, plummeting by over 6% at the opening bell the next day, serving as the most candid reflection of this rationale.

This explanation, while clear and direct, is highly likely to be misguided.

Most observers are preoccupied with the ‘AI arms race’ narrative in financial statements, completely overlooking that Google is engaged in a fundamentally distinct battle. The objective of this battle is to embed an anchor in the minds of billions. All historically successful movements that have achieved such anchoring shared a common early characteristic: financial losses were mounting, but hearts and minds were being won over.

01 The Misconstrued ‘Lag’: Unveiling the Truth Behind Model Delays

Let us momentarily set aside the tale of ‘Gemini 3.5 Pro’s arduous journey to fruition’.

What Pichai repeatedly underscored during the earnings call was not the delayed flagship model but the newly launched Gemini 3.6 Flash—a lightweight model that demonstrated double-digit improvements in programming benchmark tests while consuming fewer tokens. He also casually dropped a bombshell: Gemini 4 has commenced training, with a roadmap aiming for a pace of ‘nearly one new model per month’.

The core strategy behind these maneuvers is not to ‘construct the most formidable model’ but to ‘swiftly place a sufficiently capable model into the hands of as many individuals as possible’.

With 950 million monthly active users, 22 billion API tokens processed daily, and nearly 90% of Fortune 100 companies already utilizing Gemini Enterprise, the narrative painted by these figures is unmistakable: Google is leveraging low-cost, high-speed models like Flash as a gateway to make Gemini a habitual part of users’ lives. The delay of Gemini 3.5 Pro may be glaring on financial statements, but in the realm of user perception, it might be entirely inconsequential. With 950 million people already accustomed to incorporating Gemini into their daily routines, a three-month delay in the flagship model is merely a minor ripple in a vast lake.

This strategy mirrors how PayPal aggressively burned through subsidies to construct its payment network before the dot-com bubble burst. Wall Street persistently inquired: Where is the profit model? Peter Thiel and Musk’s response, paraphrased, was: Let us first secure 15 million wallets on our network, and we will figure out the financial aspects later.

Today, Google’s playbook remains consistent. The subsidy target has shifted from ‘registered accounts’ to ‘token consumption,’ but the objective remains to establish the default gateway for a billion individuals in the AI era.

02 Is the Money Being Squandered? First, Ascertain If What Is Being Built Is Tangible

The figure that most alarms the market—capital expenditure reaching $205 billion in 2026—is perceived as a cash-guzzling behemoth within traditional analytical frameworks. However, if we shift our perspective to ‘constructing tangible businesses with speculative capital,’ the narrative transforms entirely.

CFO Ashkenazi casually mentioned a detail during the earnings call that could easily be overlooked: approximately 60% of the quarter’s infrastructure spending was allocated to AI servers, with the remainder directed towards data centers and network equipment. She added that Google is now ‘supply-constrained,’ with demand so high that its own capacity cannot keep pace, compelling it to temporarily lease third-party cloud services.

This revelation completely alters the nature of the spending. When a company acknowledges, ‘Our capacity cannot meet demand,’ it is not gambling on an uncertain future—it is addressing a capacity shortfall on a toll highway already congested with traffic.

The billions flowing to NVIDIA and construction sites are materializing into tangible, indispensable, and immovable physical assets. This mirrors how Musk, at Tesla’s peak valuation, conducted three secondary offerings to raise $12 billion, which he promptly invested in factories in Shanghai, Berlin, Texas, and 4680 battery production lines. After the hype subsided, the factories remained, the production lines persisted, and they continue to generate positive cash flow daily.

Google is pursuing the same strategy today: capitalizing on the AI narrative’s valuation window to secure future revenue and solidify it into reinforced concrete that no one can dismantle. With Gemini boasting 950 million monthly active users and Google Cloud’s revenue surging 82% to $24.8 billion, the foundation of this fortress has already been laid.

03 The Real Battlefield Lies Not in Technology—But in a Billion Minds

This is the most counterintuitive aspect of this article.

Most individuals interpret Gemini’s 950 million monthly active users as a ‘product metric.’ However, they are misinterpreting the data. This figure actually represents the progress bar of a large-scale mindset transformation initiative.

Generative AI has been in the public eye for a mere four years. In that time, Google has acclimated nearly a billion people worldwide to conversing with an AI assistant, having it compose emails, analyze documents, and debug code. This is not technological dissemination—it is behavioral reshaping. Every interaction, every API call, reinforces a default anchor in users’ minds: When in doubt, open Gemini first.

Once this behavioral inertia solidifies in a billion minds, Gemini transcends its identity as a ‘selectable product’ and becomes a ‘default setting’ of digital life. Just as you do not say, ‘I choose to use the grid’s electricity’ today—you simply flip the switch. Whoever controls that default option possesses a perpetual money-printing machine.

This is the same trajectory Tesla followed when the global warming narrative propelled it to prominence. It did not invent any groundbreaking technology, but it dedicated two decades to delivering ideological re-education to the world, making ‘electric cars equal the future’ a pre-installed setting in a generation’s minds. By the time traditional automakers awakened, the mental real estate no longer belonged to gasoline.

Every ideological transformation in human history has initially shown no surplus in financial ledgers. 19th-century missionaries did not construct hospitals and schools worldwide by relying on profit sheets; 20th-century TV advertising giants burned through capital during prime time, with returns taking years to materialize in consumers’ minds.

Pichai described Gemini 4 as a ‘very ambitious effort’ infused with ‘massive computational power.’ Google Cloud’s revenue surged 82%. Capital expenditure guidance hit $205 billion. These figures collectively narrate a story: A civilization-level infrastructure is being deployed globally, incidentally laying down the default mental pathways for a billion people.

04 Conclusion: A Moat Deeper Than Financial Statements

The market observes negative free cash flow, model delays, and a 6% stock decline, then promptly affixes a ‘risk accumulation’ label.

However, truth seldom resides in the majority’s perception.

Google’s wager is this: When the AI era fully materializes, the default language, default mental framework, and default interactions of billions of people’s digital lives will all converge through Gemini. To secure this ‘default right,’ the first negative cash flow report in two decades is merely the commencement of the financial commitments to be made.

The arithmetic is straightforward: Once an anchor is embedded in a billion minds, the path dependency that emerges from it becomes the deepest moat on Earth. It is more enduring and harder to disrupt than any single model’s benchmark advantage.

History consistently imparts one lesson: Bubbles serve as fuel; castles are the destination. And the most critical blueprint for constructing those castles is not found on financial statements—it is the default track already laid down in a billion minds.

Analysts remain fixated on next quarter’s capital expenditure and model schedules. Google is already laying tracks for the next decade. Where those tracks ultimately lead will not be determined by Wall Street’s models—but by the 950 million people opening Gemini every day.