AI Business

Chinese Tech Giants Pour Billions Into Agentic AI Commerce as Western Rivals Pursue Different Paths

Alibaba, Tencent, and ByteDance are racing to embed autonomous AI agents directly into their e-commerce and super-app ecosystems, while Western companies pursue broader interoperability standards—a strategic divergence that could reshape enterprise AI deployment worldwide.

·5 min read
China’s hyperscalers bet billions on agentic AI as commerce becomes the new battleground
China’s hyperscalers bet billions on agentic AI as commerce becomes the new battleground

The technology sector has increasingly turned its attention to agentic AI—autonomous systems designed to complete complex, multi-step operations without human intervention. Yet this shift has unfolded along starkly different lines depending on geography. While companies in the West concentrate on developing foundational models and ensuring compatibility across multiple platforms, China's leading technology firms are pursuing a more integrated strategy centered on commerce.

Alibaba, Tencent, and ByteDance have all substantially enhanced their AI offerings to enable agentic commerce capabilities. This represents a fundamental shift away from simple conversational interfaces toward intelligent agents capable of orchestrating entire purchasing journeys, spanning product identification, comparison, and payment processing.

Alibaba recently enhanced its Qwen chatbot to facilitate transactions directly within the interface itself. By linking the AI agent to its broader ecosystem—encompassing Taobao, Alipay, Amap, and the Fliggy travel service—the company has created an environment where users can access over 400 core digital functions. Within this unified space, customers can review tailored suggestions across different services and finalize purchases without exiting the chatbot.

According to Shaochen Wang, research analyst at Counterpoint Research, "The agentic transformation of commercial services lets the maximal integration of user services and enhances user stickiness." This stronger long-term engagement, he explained to CNBC, generates durable competitive advantages.

The super app advantage

ByteDance took similar steps in December when it upgraded its Doubao AI chatbot to independently manage various functions, including booking tickets through connections with Douyin, China's version of TikTok. The company initially introduced this enhanced model on a smartphone prototype built by ZTE as a system-wide AI assistant, though certain capabilities were subsequently reduced following privacy and security objections from competitors.

During Tencent's May 2025 earnings presentation, President Martin Lau suggested that AI agents could become central to WeChat's future, given the platform's reach of more than one billion users and its integration of messaging, financial transactions, shopping, and various services.

This positioning reflects a structural advantage that China possesses in deploying agentic AI: tightly integrated platforms that avoid the fragmentation that constrains Western competitors.

Charlie Dai, VP and principal analyst at Forrester, told CNBC that "AI agents will be foundational to the evolution of super apps, with success depending on deep integration in payments, logistics, and social engagement. Chinese firms like Alibaba, Tencent and ByteDance all benefit from integrated ecosystems, rich behavioural data, and consumer familiarity with super apps."

Dai noted that Western enterprises operate within more dispersed information landscapes and face more stringent privacy rules that impede seamless service linkages. Despite their advantages in creating cutting-edge foundational models and maintaining worldwide presence, these constraints present significant obstacles.

Agentic AI's enterprise trajectory

The expansion into consumer commerce signals broader implications for business applications. As agentic AI transitions from supplementary functionality to fully autonomous operation capable of managing intricate business processes, industry observers anticipate that coordinated multi-agent architectures will emerge as a dominant pattern in how organizations deploy AI this year, extending well beyond consumer-facing applications.

Tian Feng, president of the Fast Think Institute and former dean of SenseTime's Intelligence Industry Research Institute, predicted in a Global Times report that an AI agent reaching 300 million monthly active users could materialize by 2026. Such a system would function as "an indispensable assistant for work and daily life" capable of autonomously executing cross-app, composite services.

A 2025 McKinsey study found that roughly half of all consumers currently rely on AI for online searches. The research organization projected that AI agents could unlock more than $1 trillion in economic value for American enterprises by 2030 by automating routine elements of how consumers make purchasing decisions.

Smaller Chinese cloud infrastructure providers, such as JD Cloud and UCloud, have begun offering agentic AI capabilities. However, the substantial computational costs associated with token processing have prompted certain vendors, including ByteDance's Volcano Engine, to shift toward fixed-subscription pricing arrangements to manage expenses.

Divergent deployment strategies

The gap between China's integration-focused model and the West's emphasis on scalability stems from underlying differences in how markets are structured and how regulations operate. These distinctions will probably determine which companies gain competitive advantages going forward.

Dai stated that "China will prioritise domestic integration and expansion in selected regions, while US firms focus on global scalability and governance."

In the United States, OpenAI, Perplexity, and Amazon are developing agentic commerce solutions. Google is exploring a different role—positioning itself as a facilitator connecting merchants, buyers, and AI agents. These strategies reflect the reality of fragmented platform landscapes where interoperability becomes essential rather than closed-system integration.

The autonomous character of agentic systems has nonetheless triggered regulatory scrutiny within China. When announcing Doubao's expanded capabilities, ByteDance cautioned users about potential security and privacy hazards. The company advised using the tool exclusively on separate devices rather than those holding confidential data, given the system's capacity to access device information, user accounts, and internet connections through multiple channels.

The accelerated commercialization of agentic AI across China's consumer marketplace offers business leaders everywhere valuable early insights into how autonomous systems may transform customer acquisition economics, the structure of platform competition, and the durable advantages that companies can build as these capabilities mature.