The "Three Steps" of AI Monetization: OpenAI opens the browser, Doubao provides the link, and Quark wears glasses.
The AI industry has finally reached a stage where it needs to be self - sustaining.
Recently, AI has started to "dabble in other fields".
For example, after OpenAI integrated a shopping function into its chatbox, it then ventured into browser development. Quark has developed AI glasses. And Doubao AI next door is busy adding (e - commerce) links...
Apparently, on both sides of the ocean, the fates of AI have converged, and the entire industry seems to have reached a point where it needs to support itself.
Previously, the three main approaches to AI commercialization were API calls, membership subscriptions, and customized solutions for enterprises.
Among these three approaches, the B - to - B market accounted for the largest share.
This has been a proven path for monetization. However, unfortunately, the B - end market is now even more cost - conscious than the C - end market. Model versions need to be updated, and training requires a large amount of graphics cards and electricity. The "old three" methods are increasingly insufficient to cover the huge computing power costs.
Therefore, the real money lies in the consumer - grade market.
1
Three Logics of AI Commercialization: Survival, Positioning, and Expansion
Interestingly, at this stage, the motivations of various large enterprises for exploring AI commercialization vary.
They can be roughly divided into three categories:
1. Commercializing AI based on the survival logic.
Typical examples are OpenAI and second - and third - tier domestic AI model startups.
After more and more large - scale AI model manufacturers, led by DeepSeek, have chosen the open - source route, the technological leading edge of ChatGPT and similar products has been gradually shrinking, and their external influence has also diminished significantly.
This is undoubtedly dangerous for any AI startup that still relies heavily on investors' funding.
Therefore, both internally and externally, OpenAI needs a mature business model as a value anchor to prove ChatGPT's ability to survive independently and its long - term potential.
Among all the options, search is the closest field to ChatGPT.
So, OpenAI's launch of the artificial intelligence browser ChatGPT Atlas seems, on the surface, to be an attempt to disrupt Google and recreate a Google on the basis of AI.
However, the core is the survival logic, to find a more certain path for monetization
OpenAI's choice of the browser may not be a whim. The real purpose is to create an entry point and return to centralization.
Previously, the monthly active user numbers of various large - model apps grew rapidly, and the competition was extremely fierce, resulting in an increasingly obvious trend of decentralization. In terms of user experience, instead of achieving the ideal dominance, ChatGPT has witnessed a diversified market.
Taking my personal AI usage habits as an example, I use DeepSeek for text work, Doubao for image matching, and Nano Search for complex requirements... In short, using different models for different scenarios is becoming the choice of more and more users.
This usage habit has led to the dispersion of traffic for large - scale models, making it more difficult for ChatGPT and similar products to achieve commercial success.
Therefore, OpenAI now needs to use ChatGPT to open up a traffic entry point, a centralized one with a clear commercial monetization path, to consolidate its central position in the AI ecosystem.
Technologically, AI and search have genetic similarities, and commercially, they are also highly compatible. Coupled with Microsoft's successful exploration in the field of AI browsers, these may have convinced OpenAI to enter the search field and build a sustainable commercial closed - loop...
2. Commercializing AI based on the value logic.
Typical representatives are Alibaba and other giants with AI as a long - term strategy.
Take smart glasses as an example. Why has this "old - concept product" become so popular recently?
It's not because of industrial technological progress, nor because the product form of smart glasses has become more mature. It's because of AI, specifically, natural language models.
In essence, Alibaba's Quark and others are developing smart glasses with the aim of creating an entry point.
However, different from the monetization logic of search engines, Alibaba is developing AI hardware to seize the next interaction entry point.
Alibaba has always been a giant with strong strategic planning and execution capabilities.
From its past attempts in mobile phones, to smart speakers, and now to the development of AI hardware, once Alibaba determines a strategic direction and sets the KPIs, commercial success is almost guaranteed.
The key lies in the value of AI to Alibaba. It's never just about the technology itself. Even whether AI hardware commercialization is profitable is not important. What really matters is the entry - point value represented by AI and the strategic investment value in the capital market.
This value logic has been consistent.
Previously, Alibaba ventured into new retail for offline entry points, live - streaming for e - commerce traffic entry points, and flash sales for instant retail traffic entry points. After missing out on the social media wave, Alibaba's core business model dictates that it must seize the underlying traffic entry points and ensure its position as an online "city shopping mall" for both positioning and defense purposes.
Therefore, AI today is to Alibaba what social media was in the past...
3. Commercializing AI based on the incremental logic.
Typical representatives are Douyin E - commerce and other content e - commerce players.
Currently, Doubao AI's addition of links may not only be for commercialization but also for seeking business growth.
According to data released by market research firm QuestMobile, in August this year, Doubao's MAU (Monthly Active Users) reached 157 million, a month - on - month increase of 6.6%. It not only surpassed DeepSeek to rank first but also left other competitors far behind.
With such a large traffic advantage, leaving it idle is a waste. Moreover, ByteDance missed the instant retail and local life battle in the first half of this year. Now it needs to regain the market and catch up with more traffic. Therefore, AI e - commerce represented by Doubao has become the most direct breakthrough point.
So, the logic behind Doubao's addition of links makes sense, but it's just in its early stages, so ByteDance is not taking big steps: Users still need to jump to the Douyin Mall to complete their purchases.
Fortunately, from Tieba and Zhihu in the past, to short - videos and Xiaohongshu now, and then to large - scale AI models, the level of user trust associated with different content forms has been continuously increasing.
In the future, when users really turn to AI for answers, AI will surely become the next level of user trust in transactions.
For ByteDance and Doubao, which are leading the way at present, this is undoubtedly another good opportunity to overtake competitors.
In short, at this stage, AI commercialization presents three distinct strategic maps, corresponding to different positions and demands of enterprises in the current industrial cycle:
The survival logic is "survival", which is related to the survival line of enterprises under the dual pressures of capital and market competition. Its commercialization path is often direct and urgent, aiming to quickly verify monetization ability and build a value defense line.
The value logic is "positioning", focusing on the entry point and ecological control of the next - generation human - machine interaction. Its commercialization layout is more long - term and strategic, aiming to build an insurmountable moat and ensure a core position in the future value chain.
The incremental logic is "expansion", based on catalyzing the existing large - scale traffic advantage with AI to open up new growth battlefields beyond the core business growth curve. Its commercialization exploration is more scenario - penetrative and efficiency - oriented.
These three logics are not mutually exclusive. They together outline the complex picture of AI's transition from a technological singularity to industrial integration. Now, the commercialization journey of AI has gradually evolved from a technological competition to a comprehensive contest involving survival philosophy, ecological vision, and growth wisdom...
2
Advertising and E - commerce: New "Canals" for AI Commercialization?
So far, several paths for AI commercialization have emerged at this stage:
One is the hardware carrier route: such as Quark AI glasses, AI learning machines, and Luobo Kuaipao. It targets the extremely picky consumer market. It tests the ability to provide disruptive experiences. Therefore, the corresponding market competition is relatively pure.
As long as there is a sufficiently advanced AI technology experience, it can gain consumers' recognition and become a strong brand moat.
For example, Huawei's intelligent driving technology has boosted the sales of AITO, and Luobo Kuaipao in Wuhan has achieved break - even. These are the most direct examples.
In the context where everything is worth being redone with AI, the commercial imagination on this route is almost limitless, but at the same time, it also has the most uncertainties. After all, technological breakthroughs and setbacks can happen in an instant...
The second is the software route. AI commercialization can be divided into two aspects:
On the one hand, it is the capability side, mainly including API calls, membership subscriptions, and customized solutions for enterprises. It is more about paying for AI capabilities.
For example, Keling of Kuaishou and Jimeng of ByteDance are positioned as low - threshold AI video creation tools for the C - end, and focus on three scenarios of advertising, short dramas, and games for the B - end, empowering enterprise customers by generating materials in batches and improving asset production efficiency.
This route tests forward - looking layout and the scale of industrial infrastructure. Therefore, the market pattern in some areas has initially stabilized.
For example, in the AI cloud market, Tianyancha APP shows that Alibaba Cloud's revenue accounts for 35.8%, exceeding the sum of the second to fourth - ranked companies. In terms of model call services, ByteDance's Volcengine ranked first in the number of calls to large - scale public cloud models (Tokens) in China in the first half of this year, with a market share of up to 49.2%.
On the other hand, it is the content side, mainly advertising and e - commerce, which involves inserting relevant links into AI Q&A content.
This is the simplest route, and at the same time, it has the most obvious market differentiation.
Technologically, it is easy to integrate. Different from previous industries where AI was applied, the e - commerce industry is large - scale. Once AI has a suitable monetization mechanism, the speed of technology monetization will also be faster.
However, the difference is that OpenAI (ChatGPT) is doing AI e - commerce, allowing users to complete the entire process of search (questioning) - recommendation - decision - payment in a single conversation without jumping to a third - party platform.
In this context, AI not only replaces search but also advertising. It has even become a new traffic entry point that can compete directly with Amazon for users' attention...
In contrast, Doubao's addition of links is more like an additional traffic - guiding entry point. Users still need to jump to the Douyin Mall to place orders. It still follows the traffic - driven logic, similar to the development path of live - streaming e - commerce in the past.
The impact is that when the addition of links on Doubao becomes more popular in the future, the exposure of brand products may no longer depend solely on large - scale advertising investment, but also on AI's understanding ability.
Those who can preset more marketing answers that are easily "understood" by AI will have more opportunities to reach consumers.
As for Alibaba and JD.com, if they wish, they can quickly integrate various large - scale models to achieve similar effects to Doubao and ChatGPT.
Previously, some users found that Kimi, Wen Xiaoyan, etc. have integrated shopping links from Tmall or JD.com, and clicking on them can directly lead to the e - commerce platforms.
But why don't Alibaba and others seem to attach much importance to it, and there isn't much news about it?
It's related to strategic priorities and the amount of traffic.
In the first half of this year, the absolute strategic focus of Alibaba and JD.com was the instant retail and local life battle. They made huge investments and also got considerable returns. Both sides' traffic was almost at a peak.
So now, (missing the battle) ByteDance needs AI e - commerce to stimulate traffic, while Alibaba and others don't need to emphasize it.
Also, since all companies are still in the trial - and - error stage, the strategic positioning significance of AI e - commerce is indeed important but not urgent.
Therefore, during this year's Double 11, compared to doing e - commerce within AI, Alibaba and JD.com have chosen to strengthen AI within e - commerce.
For example, Tmall has launched AI graphic designers, AI customer service, and AIGC platform tools. JD.com's AI has, for the first time, made large - scale investments in the "Super Brain + Wolf Clan" intelligent equipment cluster in the logistics system. Overall, it implements AI applications from three dimensions of matching efficiency upgrade, merchant tool popularization, and innovative product experience, with the core of "serving merchants to improve efficiency and optimizing user experience".
Apparently, different business genes have led to different AI commercialization routes.
From Quark's launch of smart glasses, to Doubao's addition of links, and then to ChatGPT's development of a browser, the wave of AI commercialization is irreversibly pushing the entire industry into a new era: the logic of value creation is being reshaped, and the core of competition is shifting from technological high - ground to ecological hinterland.
So, after this "coming - of - age ceremony", what will survive is not only the enterprises but also the new paradigms verified by the market.
This article is from the WeChat official account "Internet Jianghu" (ID: VIPIT1), author: Liu Zhicheng, published by 36Kr with authorization.