WeChat A2A Establishes Long-Term AI Advantage for Platforms

Overview

This report focuses on WeChat's launch of Agent-to-Agent (A2A) capabilities and its collaboration with China's top five smartphone manufacturers. Jefferies believes this move not only addresses current pain points in AI smartphone interaction regarding security and experience but also establishes a new paradigm for collaborative monetization between 'super apps and terminal manufacturers,' thereby arguing that large internet platforms with vast ecosystems are the long-term winners in the AI era.

Core views

Technology Route Debate: A2A is Superior to GUI Agents. The report details a comparison between two AI agent models. The GUI Agent adopted by ByteDance's Doubao Phone requires deep system permissions to simulate clicks and read screens, posing serious data privacy risks and ecosystem conflicts, leading to bans by mainstream platforms. In contrast, the A2A model promoted by WeChat enables direct collaboration between agents through standardized protocols without opening up the underlying OS, ensuring data security and control while providing a smooth and stable user experience. Although Google previously proposed a similar framework, it failed mainly because it lacked the support of a super platform with over 1 billion monthly active users like WeChat. Reshaping the Business Landscape: Win-Win for Smartphone Manufacturers and Platforms. For smartphone OEMs, A2A upgrades them from mere 'app shelves' to 'user intent routers.' As more applications connect, smartphone manufacturers will gain new bargaining power in distributing user demands, allowing them to share revenue from online merchants without significant hardware upgrades. For WeChat, A2A allows it to connect with over 30,000 AI mini-programs (such as Meituan, JD.com, and KFC) via in-app agents without handing over core social and financial data, significantly boosting in-app transaction volumes and fee income, truly evolving towards an 'AI Operating System.' Industry Endgame Judgment: Internet Platforms are Long-Term AI Winners. The report points out that other super apps like Alipay and Taobao may follow suit to build A2A ecosystems. Large internet platforms possess rich C-end AI monetization scenarios, and this revenue can, in turn, fund their B-end AI R&D. This 'C-side supporting B-side' closed-loop capability makes them more likely to survive and win in the long run compared to pure AI labs lacking commercialization scenarios. Conversely, products relying on GUI Agents, such as Doubao Phone, face dim prospects for scaling due to the lack of support from head platforms.

Analysis framework

The report adopts a typical technology-business dual-drive analysis framework. It first deconstructs the essential differences between A2A and GUI Agents at the technical principle level, demonstrating the structural advantages of the former in terms of security, stability, and ecosystem compatibility. It then maps these technical advantages to the benefit distribution mechanisms across the industry chain, analyzing the incentive compatibility among smartphone manufacturers, super apps, and third-party developers. Finally, combining the unique 'super app' ecosystem barriers in China's internet sector, it deduces the moats of platform companies in the long-distance race of AI commercialization. This analytical method avoids merely discussing technical parameters and instead focuses tightly on the capital market's core concern: 'Who can make money.'

Methodology notes

• Competition and Strategy FrameworkMoat / competitive advantage

  Super app ecosystem barriers as the core moat for AI monetization

  The report emphasizes that WeChat's over 1 billion monthly active users and 30,000+ mini-program ecosystem are assets difficult for pure technology companies to replicate. In the AI era, this network effect and user stickiness constitute a deeper moat than algorithms themselves, determining who can translate technology into actual revenue.

• Industry/Sector Analysis FrameworkUpstream-Midstream-Downstream Transmission

  Value redistribution in the industry chain triggered by changes in AI agent models

  The report analyzes how A2A changes the relationships between smartphone manufacturers (upstream), super apps (midstream), and service providers (downstream). Smartphone manufacturers shift from hardware sales to service revenue sharing, while super apps transition from traffic entry points to intent distribution hubs. This value chain restructuring is the key logic for judging the degree of benefit for each party.

• Competition and Strategy FrameworkSubstitution Effect Analysis

  Technical substitution and ecosystem crowding-out of GUI Agents by A2A

  By comparing the security and compatibility of the two technical routes, the report argues why A2A will become the industry standard. When a technical solution can simultaneously meet platform security demands and user experience needs, another solution with structural defects will be naturally eliminated by the market. This is the core basis for judging the dim prospects of Doubao Phone.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Tencent (WeChat)

  Leader of the A2A ecosystem, directly benefiting from AI transaction monetization

  Strengths

  Possesses 1 billion+ monthly active users and a 30,000+ mini-program ecosystem, controls core social and payment data, and has the ability to define industry standards

  Comparison

  Compared to ByteDance, Tencent chose a more developer-friendly open protocol route; compared to pure AI labs, it possesses an existing commercial closed loop

• Smartphone OEMs (Huawei/Xiaomi/Honor/OPPO/vivo)

  A2A participants, benefiting from role transformation and revenue sharing

  Strengths

  Control terminal entry points, can participate in AI monetization without major hardware upgrades, and have enhanced bargaining power

  Weaknesses

  Dependent on the willingness of super apps like WeChat to open up; relatively weaker internal AI capabilities

  Comparison

  Compared to Doubao Phone, the A2A model does not require manufacturers to open up the underlying OS, resulting in less resistance to cooperation

• ByteDance (Doubao Phone)

  Representative of the GUI Agent route, facing risk of marginalization

  Weaknesses

  GUI Agents pose security risks and ecosystem conflicts, are banned by mainstream platforms, and are difficult to scale

  Comparison

  Lacks ecosystem support from head internet platforms compared to the A2A camp

  Risks

  Products may gradually fade from the market

Key data

• WeChat Monthly Active UsersOver 1 billionCore foundation supporting the scale effect of the A2A ecosystem, and a key differentiator from the unsuccessful Google AppFunctions
• Number of AI Mini-Programs30,000+Including Meituan, JD.com, KFC, etc., constituting the service supply pool for WeChat's AI OS
• Smartphone Manufacturers in A2A Cooperation5Huawei, Xiaomi, Honor, OPPO, vivo, covering China's mainstream Android camp

Impact & implications

The report believes this development has profound implications for China's tech sector. For Tencent, A2A opens up new spaces for transaction commissions and advertising monetization, reinforcing its position as an AI infrastructure provider. For the smartphone supply chain, hardware manufacturers have gained new growth points through software servitization, which may improve low-margin hardware business models. For the entire AI industry, this marks a shift in competitive focus from pure model capabilities to ecosystem integration and commercial closed loops, challenging the valuation logic of pure AI companies lacking application scenarios. Meanwhile, platforms with similar ecosystems, such as Alibaba, may replicate this path, forming a multi-polar landscape of AI operating systems.

What to watch

• Whether other super apps (such as Alipay, Taobao) follow suit with A2A frameworks
• Actual adoption speed and activity levels of third-party mini-programs in the A2A ecosystem
• Implementation details of specific revenue-sharing mechanisms for smartphone manufacturers under the A2A model