According to a research report, Alibaba-W's (09988) core AI strategy is no longer centered on the scale or benchmark performance of a single model. Instead, the company is systematically integrating its "Tongyi Qianwen" model with high-frequency scenarios within its digital economy ecosystem, such as e-commerce, local services, payments, transportation, and office collaboration. This integration aims to create a closed-loop user experience characterized by the ability to "get things done" effectively. Furthermore, Alibaba is elevating AI from being a supplementary tool to becoming the primary interface for users to access services, effectively making "AI the entry point."
Alibaba's overarching AI strategy has shifted from competing on individual models to competing with a holistic system. This approach leverages the integration of the Qianwen interface with the collaborative Cloud-DingTalk-Bailian platform, alongside increased capital expenditure on cloud and AI infrastructure. The objective is to build closed-loop capabilities to compete for dominance in the next-generation platform landscape.
The key viewpoints from the report are as follows: Alibaba's top-level strategy is evolving. The focus has moved from "comparing models" to "competing with systems." The company is utilizing a combination of "models + ecosystem + AI infrastructure" to vie for leadership in the next-generation platform.
Post-2025, the core of Alibaba's AI strategy emphasizes the systemic connection between Tongyi Qianwen and its high-frequency digital economy scenarios. This aims to create a practical, closed-loop experience and position AI as the main service access point.
Internally, Alibaba has introduced the "Tongyun Ge" concept, which binds the Tongyi Lab, Alibaba Cloud, and T-Head together as an integrated "golden triangle." This emphasizes the synergy between computing power supply, model capabilities, and systems engineering, attempting to build a universal foundation for the AI era from the infrastructure level.
The strategy targets both consumers and businesses. For consumers, the focus is on user scale and super-applications. For businesses, it targets commercialization. The Qianwen App's rollout as a primary interface is based on its "task execution capability and ecosystem-level connectivity," establishing its differentiation.
A significant upgrade in January 2026 saw the app fully integrated with ecosystem services like Taobao, Alipay, Taobao Flash Sales, Fliggy, and Amap. This marks Alibaba's move from conceptualizing to actively implementing its "entry point battle" for consumer users.
DingTalk is positioned as an "enterprise-grade AI agent platform," tasked with embedding model capabilities into corporate workflows and business systems. On the supply side, the integrated "Cloud-DingTalk" strategy and the Bailian platform focus on the large-scale deployment of AI agents. The business model is shifting from selling functional subscriptions to selling outcomes and productivity.
Alibaba is clearly directing substantial capital investment towards cloud and AI infrastructure. This includes expanding data centers, server clusters, and the procurement and deployment of high-performance GPUs. The company is also advancing the in-house development of cutting-edge AI chips through T-Head to support model and platform expansion.
On the capacity side, capital is being rapidly converted into tangible computing power supply. Efforts include advancing the construction of global data centers and nodes, expanding existing facilities, and adding AI-dedicated suites of hardware, all serving its path towards a "super computing cloud platform."
Risk warnings include potential slower-than-expected development of AI technology, challenges in the commercial implementation of AI, and intensifying market competition.