As red packets once again become the "nuclear button" for Chinese internet companies, this time it is not payments being ignited, but AI assistants. According to market observations, with the 2026 Spring Festival approaching, the battle for traffic among Chinese internet giants surrounding AI assistants is rapidly heating up. Tencent and Baidu have taken the lead in announcing large-scale cash red packet subsidies, while Alibaba and ByteDance are widely expected by the market to join the fray. A battle for the AI entry point, using "Spring Festival red packets" as the trigger, has quietly begun. Citigroup points out in its latest research report that this is not merely a simple Spring Festival promotion, but rather resembles a critical validation window for Chinese AI assistants approaching an inflection point towards mass adoption. From "user acquisition via red packets" to "the battle for AI entry points," the underlying logic has fundamentally changed this time. Spring Festival red packets are nothing new. From WeChat Pay and Alipay to short video platforms, over the past decade, almost every round of internet traffic restructuring has been accompanied by Spring Festival subsidy wars. However, Citigroup emphasizes that the key difference in 2026 lies in the fact that the core object of contention has upgraded from "applications" to "the traffic entry point for the AI era." In Citigroup's view, AI assistants possess three characteristics that previous applications lacked:

Naturally high-frequency: Search, Q&A, and content generation create usage scenarios that are extremely fragmented yet highly daily. Strong platform attributes: Once established as the default entry point, they can continuously direct traffic to e-commerce, advertising, cloud services, and finance. Ecosystem amplifier: AI does not monetize independently but rather amplifies the efficiency and stickiness of the entire ecosystem.

This is also why Citigroup defines 2026 as "the year of full-scale escalation in the traffic battle for AI chatbots." The Spring Festival red packet is merely the form; what is truly being tested is the "formation of usage habits." Judging from the disclosed information, the strategies of various players have clearly diverged:

Tencent: Its assistant Yuanbao announced a 10 billion RMB cash red packet investment, with the core pathway being "try AI functions - draw rewards," emphasizing high-frequency interaction and social diffusion. Baidu: Investing 5 billion RMB in red packets, it has deeply embedded its Wenxin assistant into the Baidu App, guiding users to repeatedly use it through "AI interactions + card collection gameplay." ByteDance: While not directly announcing a red packet amount, it secured Doubao massive exposure in a "national-level scenario" by making its Volcano Engine the exclusive AI cloud partner for the CCTV Spring Festival Gala. Alibaba: Although not officially announcing Spring Festival red packets yet, it previously validated Tongyi Qianwen's content interaction capabilities during Bilibili's New Year's Eve show, making a Spring Festival follow-up highly probable.

Citigroup points out that the essence of the Spring Festival red packet is not the subsidy amount, but "forcing users to form AI usage pathways." As long as users develop reliance through 7-10 days of high-frequency use, post-holiday retention rates could be significantly higher than those of historical internet products. The data is already speaking: AI assistants are approaching the threshold of "mass adoption." Looking at存量 data, the user scale of Chinese AI assistants is no longer at an "experimental level." According to QuestMobile and MoonFox data cited by Citigroup, as of December 2025:

Doubao's MAU reached 227 million, firmly holding the top position. DeepSeek, Yuanbao, Tongyi Qianwen, and Wenxin assistants had approximately 136 million, 41 million, 26 million, and 5 million MAUs respectively. Doubao's DAU had already reached 70 million, with the average usage time per user for some AI assistants exceeding 100 minutes.

More notably, the proportion of Web-based DAU has increased significantly for multiple AI assistants. For Baidu's Wenxin assistant, Web usage accounts for as much as 65%, indicating that AI is transitioning from an "App tool" to a "universal entry point." This is the key judgment Citigroup repeatedly emphasizes: the competition for AI assistants has shifted from "model capability" to "who can become the default entry point first." Will the "iPhone Moment" truly arrive? So the question arises: After the Spring Festival, will China's AI assistants truly experience their own "iPhone moment"? Citigroup's answer is relatively measured but clear in direction:

The Spring Festival red packets will not directly determine the winner. However, they will constitute the first nationwide stress test with real user density. The ability to maintain high DAU and high usage frequency after the holiday will become the watershed separating "tool-based AI" from "platform-level AI."

In terms of investment preference, Citigroup's ranking based on "traffic entry point + ecosystem synergy" is: Tencent > Alibaba > Baidu, a judgment that also resonates with recent stock price performance. Conclusion: Red Packets are the Starting Point, Not the Finish Line Looking back, every generational leap in the Chinese internet has had a "mass adoption引爆点":

The 2014 WeChat red packet brought mobile payment. The 2018 short video subsidy shaped a new paradigm for content distribution. And the 2026 Spring Festival might be regarded as the starting point for Chinese AI assistants transitioning from "usable" to "essential."

The red packets will fade away, but once user habits are reshaped, the real competition will have just begun.