11/17 2026
353
I believe many of you still vividly remember the AI red envelope battle during the Spring Festival. A key reason players were willing to "burn money" on AI is that this approach allowed them to rapidly acquire traffic and thereby secure an advantageous position in the new AI battleground.
After the Spring Festival AI red envelope battle, the competitive landscape among players has undergone profound changes, and those with abundant cash reserves have begun to stage a comeback.
As AI development continues to deepen—especially now that traffic competition can no longer be won simply by "spending money"—more and more players are turning their focus to rich and diverse scenarios.
How to apply their AI tools to as many scenarios as possible, genuinely help scenarios and industries solve real-world problems, and thus sustain their advantages in the new AI battleground has become a new choice for a growing number of players.
In the past, during the traffic competition era, players competed more over how to make more of their traffic usable. Now, in the scenario competition era, players compete more over how to make more scenarios and industries renewed because of AI. More precisely, how to make AI evolve from "can chat" to "can work" is becoming a new choice for more and more players.
Traffic Is Only the Beginning; Commercialization Is the Endgame
In the mobile internet era, we actually witnessed the same scene: players would acquire traffic by "spending money" and quickly occupy an advantageous position in the market. Once that advantageous position was established, they would seek new growth methods more from the perspective of commercialization.
In fact, in the AI era, the same scene is playing out.
For example, during the Spring Festival, we saw a "money-spending battle" among players. However, having traffic is only the beginning. How to apply AI to more scenarios—and, after completing traffic cultivation, provide new products and services from the supply side to the traffic that has already been cultivated—has become an important issue that players must inevitably consider.
Take Shenzhou Holdings' Kejie as an example. Before its supply chain intelligent agent "Jie Xiaoguo" went online, the team spent four months going through 118 supply chain positions, sorting out 115 requirements, and organizing the experience of senior employees into more than a hundred directly callable skills. During "Double 11," employees from 13 business departments actually used the intelligent agent, completing 14,000 rounds of dialogue and invoking tools 95,000 times, with an average of 6.8 invocations triggered per question. Every time an employee asks a question, the intelligent agent enters the system to execute nearly seven queries and calculations. In the past, retrieving data across systems required submitting a request first and then waiting for scheduling. Now, one sentence can directly connect to multiple business systems to obtain reports. This is precisely a typical path of reshaping production methods with AI from the supply side after traffic cultivation is completed.
Therefore, the competition for traffic is only the beginning for AI players. How to apply AI to as many scenarios as possible, bring about the transformation of industries and scenarios through the commercialization of AI, thereby achieving transformation and upgrading on the supply side, ultimately connecting both supply and demand, and building a complete commercial closed loop—this is the important issue players must inevitably consider.
From this perspective, "can chat" based on traffic is only the beginning, while "can work" based on commercialization is the key to truly ensuring that AI players can continue to maintain leadership in the future.
A Single Scenario Is Only the Beginning; Scenario Collaboration Is the Key
As AI begins to enter a brand-new stage centered on applications, players in almost every scenario will have AI applications unique to their own scenarios.
For example, in the education scenario, there are AI tools exclusive to education; in the food delivery scenario, there are AI tools exclusive to food delivery; in the ride-hailing travel scenario, there are AI tools exclusive to ride-hailing... In other words, AI today has long evolved from being exclusive to only a few players into a standard configuration for almost every player. The "Workbench" function recently launched by Doubao Work covered 21 professional scenarios in its first batch, including finance, healthcare, education, and creative design. Doctors can use it to find clinical guidelines and search medical literature, while teachers can generate lesson plans and courseware with one click—or even turn a question into an interactive experiment with draggable parameters.
In fact, this is the inevitable result of AI moving toward deep application.
Only by developing applications unique to each scenario can AI be quickly and deeply integrated with scenarios and industries. However, it must also be noted that AI applications in a single scenario are only the beginning.
Vertical AI tools for a single scenario can only solve problems in that single scenario. To maximize AI's functions and effects, it is necessary to establish deep collaboration among different scenarios and industries to achieve greater results. The path of Alibaba's Qwen provides a reference: Qwen has been fully integrated into core businesses such as Taobao, Alipay, Amap, and Fliggy, attempting to connect e-commerce, local life, and financial scenarios through a unified AI interface, forming an end-to-end closed loop from life services to payment fulfillment. Users do not need to jump between different applications; a single entry point can mobilize cross-scenario service capabilities.
The implementation and application of a single scenario are only the beginning. What it achieves is the renewal of a single scenario and industry based on AI. To maximize the effects of renewing different scenarios and industries with AI, it is necessary to achieve deep collaboration among different scenarios and industries, so as to find more new development possibilities.
From this it can be seen that AI applications under a single scenario only achieve the upgrading of a single scenario. To truly unleash AI's effectiveness, deep collaboration among scenarios and industries is still needed, rather than merely forming one isolated island after another.
Scenario Transformation Is Only the Beginning; a New Balance Is Fundamental
When players begin to focus on the implementation and application of AI in scenarios and industries, the focus of their competition shifts to who can implement AI in as many industries and scenarios as possible and who can make as many scenarios and industries as possible achieve comprehensive renewal.
In this way, different types of scenarios and industries can indeed achieve comprehensive renewal, but merely achieving the renewal of scenarios and industries is still only the beginning.
For players, how to build a balance between a supply side based on scenarios and industries and a demand side that has already completed AI cultivation, and form a complete commercial closed loop, is the key to truly ensuring the perfect implementation of AI.
The practice in Anqiu, Shandong, provides an observation sample from small and medium-sized enterprises. After a local silk flower handicraft factory connected to an industrial intelligent agent platform, overseas customers submit customization requests online, and the system can generate high-definition 3D proofs in a few seconds. After customers make online fine-tuning, it automatically quotes, splits orders, and schedules production. The process that previously required at least half a month of drawing blueprints and mailing samples back and forth has been compressed to on-the-spot confirmation. The same product with personalized customization is priced at more than double that of standard products, and comprehensive profit has increased by more than 20%. Here, what AI brings is not only efficiency improvement on the production side, but also the establishment of a new balance between supply capacity and personalized demand.
From this point of view, the implementation of AI in scenarios and industries is only the beginning. It only enriches the scenario side and realizes the transformation from "AI+C-end" to "AI+B-end."
To truly combine the scenario-based implementation of AI with real commercialization, it is necessary to build a new balance between the supply side and the demand side based on AI.
When such a supply-demand balance is established, the scenario-based implementation of AI will no longer be merely an answer without results, but a closed loop with real user demand and willingness to pay.
It is not difficult to see that shifting from C-end to B-end is only a transfer of the transformation target and only the beginning of AI exerting greater efficiency. Truly building a new balance between supply and demand is the key to truly ensuring the success or failure of AI commercialization.
In Conclusion
Now, more and more players are beginning to evolve their AI from "can chat" to "can work," and some players have even begun to embed AI tools deeper into more and more industries and scenarios.
This is inevitable in AI development. Essentially, only by realizing the implementation of AI in scenarios and industries can AI's roots go deeper. With the implementation of AI in scenarios and industries, a new landscape will emerge, and new directions will also surface.