At the critical juncture where the AI industry transitions from technological fervor to commercial implementation, the “reversal of the application ecosystem dilemma” has become the key to breaking the deadlock. Following the emergence of the pattern of “West Google, East Alibaba,” marginal changes in the industry are confirming a trend:AI infrastructure’s massive investment will ultimately spur the profit inflection point of ecological applications, while the layout of Eastern and Western tech giants is reconstructing the growth logic of the globalAI industry.
1.AI‘s essential bubble is capital, but the ecological dividend remains unextinguished
Recently, fluctuations in the US stock market have often been labeled as a “bubble”: expectations of interest rate cuts by the Federal Reserve are wavering, and the Bank of Japan’s balance sheet reduction has triggered capital withdrawal… Under multiple factors,M7 tech giants’ valuations are under pressure. However, the uniqueness of the AI track lies in the fact that the noise of capital cannot obscure the inevitable returns from technological investment.
NV earnings exceeded expectations, and Google’sGemini3 product capabilities have broken through (topping theLLAMA arena), dispelling doubts with performance:AI hardware has long proven its profitability, and the profit release of ecological applications is merely a matter of time. Just like “there’s no need to question the intelligence of the wealthy,” the tech giants’ investments inAI are by no means blind; when the “hard power” of computing infrastructure is completed, the monetization of the “soft power” of the ecological closed loop will become the main theme of the next stage—scenarios where data is organized usingExcel and payment is made via QR code, generatingPPT, essentially reflect the penetration of ecological value into the C-end.
2. Google’s ecosystem: The “closed-loop dominance” of the hexagonal warrior“
If we talk about the global benchmark for the AI ecosystem, Google can be regarded as the “hexagonal warrior.” The product capability iteration of itsGemini series models (such asGemini3 Pro) not only showcases technological research and development barriers but also relies onGoogle Search,Google Workspace,YouTube, and other services to build a closed-loop ecosystem of “input-processing-output-monetization”: user data feeds back into model optimization, model capabilities empower product experiences, and paid services (such asCloud AI, value-added tools within the ecosystem) achieve a commercial closed loop.
This ecosystem’s “blood production capacity” is directly reflected in stock prices and financial performance: stable cash flow supports continuous investment, and the closed-loop ecosystem withstands cyclical fluctuations, allowing Google to navigate independently through the turbulence of tech stocks. For investors, the term “ecosystem” is not only an aesthetic consensus but also a benchmark for performance certainty—powerful as Apple, its high gross margin in service business also stems from ecological stickiness, while Google is pushing the story of “AI + ecosystem” to new heights.
3. Alibaba’s ecosystem: The “benchmark” and “disruptor” of the Eastern track“
Under the narrative of “West has Google,” the industrial logic of “East has Alibaba” is equally solid. As the world focuses onAI infrastructure, Alibaba centers around the “QWEN” chain, anchoring the threefold inflection point of “hardware profitability + ecosystem advancement + application revenue generation”:
On the hardware side:NV‘s quarterly report shows $57 billion in revenue, and Baidu’sAI native marketing services reached 2.8 billion (up 262% year-on-year), confirming the commercialization validation of the tech giants’ investments inAI infrastructure;
On the ecosystem side: Alibaba’s e-commerce, cloud computing, and local life form a super ecosystem, providing natural scenarios forAI applications (such as content generation on Taobao, intelligent scheduling in Cainiao logistics);
On the application side: From lightweight scenarios like “Taobao PPT generation” to industrial-gradeAI solutions, the path of the ecosystem feeding back into applications is becoming increasingly clear.
If Google has proven the “monetization efficiency of overseas ecosystems,” Alibaba has demonstrated the “depth of domestic ecosystem scenarios”—whenAI shifts from a technical competition to a commercial efficiency contest, whoever can bind the broadestC end users andB end demands will hold the pricing power of the ecosystem.
4. The “restless cycle” of AI applications“
From a temporal perspective, the “dilemma reversal” of the AI industry is brewing: hardware revenue solidifies the foundation, ecosystem advancement opens up growth space, and application explosions ignite valuation elasticity. From the fourth quarter of 2024 to 2026, this cycle switch from “hard to soft, from infrastructure to ecosystem” will become the core clue for tech stock investment.
Google’s Gemini/OCS chain: The technological spillover dividend of a globalized ecosystem
1. Zhongji Xuchuang (300308.SZ):Global leader in AI optical modules, bound to Google’s supply chain
Core supplier for Google’s data center construction,the shipment ratio of 800G/1.6T optical modules exceeds30%. As the demand for computing power in theGemini ecosystem explodes, the expansion of AI server clusters will directly drive the demand for high-speed optical modules.
Technical barrier: Mastering core silicon photonics integration technology, with industry-leading yield;
2. Shenghong Technology (300476.SZ): Pioneer in high-end PCB boards, deeply embedded in the AI server supply chain
Core PCB supplier for Google’s TPU chips,HDI boards and high-speed multilayer board products are certified by Google, accounting for about25% of the cost of a single AI server PCB.
Process barrier:HDI boards have a minimum line width/spacing of20μm, meeting the high-density interconnection requirements of AI chips;
Customer stickiness: Over5 years of cooperation with Google, participating in the research and development of the new generation TPU architecture, with order cycles lasting up to3 to 5 years;
3. Tengjing Technology (688195.SH):Invisible champion of laser radar optical components, benefiting from AI terminal hardware innovation
Core optical component supplier for Google’s Waymo autonomous driving laser radar, with a market share of over60% for narrow linewidth lasers, collimating lenses, and other products.
Technical scarcity: Mastering ultra-fast laser processing technology, breaking foreign monopolies;
Scenario extension: The demand for optical sensing in emerging hardware such as AI glasses and robots is surging, with the global market for AI terminal optical components expected to reach$4 billion by 2025;
Alibaba’s QWEN chain: The localized ecosystem’s deep cultivation barrier
1. Data Port (603881.SH):Leading third-party data center IDC, core service provider for Alibaba Cloud
Operating over50% of Alibaba Cloud’s data centers nationwide, providing customized cabinet hosting and operation maintenance services, accounting for over40% of Alibaba Cloud’s IDC costs.
Resource barrier: First-tier cities have over100,000 cabinets in IDC reserves, with PUE values as low as1.15, meeting the high computing power demands of AI;
Policy catalyst: The “East Data West Calculation” project accelerates the construction of western nodes, with Data Port laying out capacity releases in Inner Mongolia, Ningxia, and other regions.
2. Worth Buying (300785.SZ):Leader in consumer content AI, integrated into the Taobao content ecosystem
Exclusive partner for AI content generation tools in Taobao’s “Browse” section, providing users with personalized product recommendations and graphic creation services.
Scenario scarcity: Covering500 million monthly active users on Taobao,AI generated content click-through rates are three times higher than traditional recommendations;
Monetization model:C end subscription service (9.9 yuan/month for AI shopping guide) + B end advertising revenue sharing (15% of GMV) driving dual engines;
Data feedback: After using the AI tool, users’ average transaction value increased by20%, boosting the monetization efficiency of Taobao’s content ecosystem.
3. Runze Technology (300442.SZ):Expert in liquid cooling temperature control solutions, bound to Alibaba Cloud’s green data centers
Providing immersion liquid cooling systems for Alibaba Cloud’s Zhangbei data center, reducing energy consumption by40%, with PUE values ≤1.08.
Technical exclusivity: The only company in China to achieve mass production of liquid cooling heat dissipation for single chips over10kW;
Policy dividend: Under the “dual carbon” goal, the market size for energy-saving renovations of data centers exceeds$50 billion, with Runze’s market share exceeding20%;
The “freezing point” of the AI industry has never denied the value of technology, but has instead forced the maturation of ecological logic. The pattern of “West has Google, East has Alibaba” is not only a reflection of global tech competition but also a Chinese-style interpretation of the commercial closed loop of “infrastructure-ecosystem-applications.”
The investment logic of the Google chain and the Alibaba chain essentially represents the parallel tracks of “global technology premium” vs. “localized scene deep digging.” The former benefits from the overseas expansion of AI infrastructure and the arms race for computing power, while the latter relies on China’s largest consumer internet ecosystem to achieve cost reduction and efficiency enhancement.
(The above content is based on publicly available information and does not constitute investment advice.)
Luojia Investment-zX0077
Investment Advisor:Lan Yongyuan
Qualification Number:A0790624100013
Risk Warning: The stock market has risks, and investment should be cautious. Any operations based on this are at your own risk!