The full integration of Qianwen means that the paradox concerning Taobao has only just begun.
```
Author | Song He
In November 2019, Disney made a decision that excited Wall Street:
Launching Disney+, going head-to-head with Netflix.
At that time, Disney was one of the most profitable media companies in the world, but the rise of Netflix created intense narrative pressure: cable TV was in decline, users were leaving, and unless Disney seized the entry to streaming, its content would be relegated to a supplier on other people's platforms.
So Disney pulled its content from Netflix and invested heavily in creating its own streaming service. Subscriber numbers soared into the hundreds of millions, but the cost started to become clear:
Traditional cable TV business shrank faster, ESPN subscribers lost even more quickly, the window for theatrical releases was compressed, and streaming itself suffered huge annual losses.
Disney+ succeeded in user metrics, but Disney's overall profit margin actually decreased. Bob Iger was brought back largely to deal with this situation.
The direction of the new business was not wrong. But the profits of the old business were accelerated by the new business itself.
Sometimes, taking action is necessary, but effort does not necessarily make things better.
Alibaba in May 2026 faces a structurally similar challenge.
On May 13, Alibaba released its Q4 results for fiscal year 2026, showing that external commercialization revenue for Alibaba Cloud accelerated to 40%, AI-related product revenue reached 8.971 billion yuan, maintaining triple-digit year-on-year growth for eleven consecutive quarters, with annualized revenue exceeding 35.8 billion yuan. AI accounted for more than 30% of cloud external revenue for the first time.
Core e-commerce CMR comparable growth was 8%. MaaS platform BaiLian saw customer numbers up 8x year-on-year. T-head’s self-developed GPU chips are in mass production, with 60%+ computing power serving external clients.
Wu Yongming confirmed during the conference call: “Alibaba’s full-stack AI technology investment has officially crossed the initial incubation stage and entered a positive cycle of large-scale commercialization returns.”
These numbers are real and solid, and Alibaba has become the first internet giant in China to deeply integrate large models with the e-commerce ecosystem.
On May 11, Qianwen and Taobao announced full integration; it took only half a year from Qianwen’s public beta to full access, involving Taobao’s 4 billion product library, Cainiao, Flash Sale, and the entire ecosystem.
Continuing to invest amid a net outflow of 46.6 billion yuan in annual free cash flow requires determination; but if you only see courage and speed in this storyline, you might overlook some critical details.
1. Three Layers of Anxiety in AI E-commerce
What would happen if Taobao didn’t connect Qianwen?
In the short term, probably nothing drastic.
Taobao's monthly active users are at 951 million, CMR is still growing, 88VIP members and active buyers maintain double-digit growth. The basics are stable.
Currently, the impact of AI shopping on e-commerce structure is more in narrative. For example, Doubao’s e-commerce internal testing just started volume, JD AI Shopping is still early, Tencent Yuangbao is positioned as a decision support tool rather than a closed transaction loop. No AI application has truly grabbed large-scale GMV from Taobao.
But Alibaba's anxiety is real and multi-dimensional.
The first layer of anxiety comes from capital narrative.
Over the past two years, Alibaba has told the market a clear AI story: 380 billion yuan in infrastructure investment (Wu Yongming later said this plan may be underestimated), Qianwen App with 300 million MAUs, BaiLian customers up 8x.
This story has already been priced in by the market, and Alibaba's valuation logic has shifted from "e-commerce profit-driven" to "AI + cloud growth-driven".
If the AI narrative lacks commercial evidence, the valuation system will be shaken.
Qianwen's integration with Taobao is essentially providing the most intuitive, grounded scenario for the current story: showing the market that AI can really help Taobao sell goods.
The second layer of anxiety may come from competitors.
Doubao’s 345 million MAU is firmly first among domestic AI apps; in March 2026, it went into internal testing with Douyin E-commerce for "one sentence shopping". JD launched independent JD AI Shopping app at the end of 2025. Tencent Yuangbao's daily active users topped 50 million during the Spring Festival. Kimi has integrated product jumps for both Taobao and JD.
In this atmosphere, not participating can easily make the market think “Alibaba is falling behind in AI e-commerce.”
The third layer of anxiety may come from muscle memory.
In the past three years, Pinduoduo and Douyin E-commerce eroded Taobao’s market share from price and content, respectively, with CMR growth near zero or negative for several quarters.
Since fiscal year 2022, Alibaba has experienced a painful period of stall: in FY2023 CMR dropped sharply, not returning to growth until FY2024. This left a muscle memory: can't wait until threats take shape, must enter early.
The urgency of Qianwen’s integration into Taobao stems partly from rational judgment about AI entry, and partly from an instinct of "never be a step too slow again."
With all three layers of anxiety superimposed, “Qianwen’s full integration into the e-commerce system” is likely a fast and forceful strategic decision driven by such a background.
Direction is likely not wrong; the integration of AI and e-commerce is a deterministic trend, but ensuring the direction, timing, rhythm, and form are all correct is a high-difficulty move.
2. The Paradox of Efficiency and Dwell
The structural contradictions encountered by Qianwen’s integration into Taobao do not disappear just because the direction is correct.
Taobao’s commercial model over the past twenty years can be condensed into one sentence:
A user's time is the platform’s inventory.
Shopping on Taobao, price comparison, watching livestreams, scrolling recommendations, reading reviews, adding to cart—these behaviors are not necessarily required for placing an order, but they are Taobao’s most core weapons.
Ad exposure relies on dwell, recommendation conversion relies on browsing depth, impulse buys rely on each extra second spent. Taotian Group’s latest quarterly CMR comparable growth was 8% in FY2026; over 100% of net profit comes from customer management revenue—other businesses combined are still loss-making.
CMR is Alibaba's cash source, and one of its engines is precisely "how much time users spend on Taobao".
AI shopping pursues the opposite: faster task completion, narrowing choices, shortest path to order.
Qianwen's product logic is clear: user says one sentence, AI finds products meeting six criteria, thirty seconds replaces twenty minutes of decision-making.
A platform makes money from users wasting time, now Alibaba has made a tool to help users save time.
This contradiction is very similar to Disney’s dilemma: the better Disney+ performs, the faster ESPN and theaters lose users; the better Qianwen shopping works, the thinner is Taobao’s information flow ad exposure base.
Of course, Alibaba has exercised restraint in integrating the two—for example, at this stage Qianwen does not prioritize ad products, showing Alibaba’s choice of “build experience first, monetize later”. But it also means it hasn’t figured out how to make money from ads in the AI shopping flow.
Traditional e-commerce models are built upon some degree of information asymmetry, while AI is naturally a tool to eliminate that asymmetry.
When a user says: "Help me buy the best ergonomic chair", AI recommends the best product, ad system fails; recommend the highest bid product, user trust collapses.
This is likely the structural tension all future e-commerce platforms must face between efficiency tools and the attention economy. Alibaba may encounter this problem earlier than others precisely because it’s pushing the furthest.
The other side of user dwell is a new challenge for the ad system.
Taobao’s current ad system is the result of twenty years of evolution; full-site promotion penetration is still rising, 30% target for FY2026, next year’s goal raised to nearly 50%—clearly a key CMR growth driver.
But AI shopping interaction may fundamentally change traffic allocation logic; once users shop via conversation rather than browsing, "exposure" no longer applies.
Replacing Taobao’s infinite scroll product feed with AI-recommended three to five products is a difficult transition, especially since the old system contributes hundreds of billions in CMR revenue each year.
Even if Alibaba Mama lands AI Wanxiang to address this change, replacing a profitable old system with a new but unproven one still carries risk.
Recall Disney’s lesson: Disney+ burned billions on content, subscriber numbers soared, but high-profit cable ad revenue disappeared faster. New business revenue growth did not outpace old business profit decline.
Taobao’s future challenges may not be identical, but the logic is similar; the key is whether the incremental value from AI shopping can offset its erosion of the traditional ad system.
3. Mismatched Scenarios for Strength
In terms of interaction, there are fundamental differences between AI and traditional e-commerce models.
AI shopping is better suited for standardized products, especially SKUs with low quantity, transparent pricing, and no emotional decision—like tissues, data cables, cat food, batteries.
Spring Festival data supports this: 130 million users experienced AI shopping for the first time, nearly half the orders came from county-level areas, nearly 4 million users aged 60+ placed orders with “one sentence.”
This shows AI does have huge value in lowering thresholds and reaching new users.
But this scenario for standard products is closer to Pinduoduo’s comfort zone than Taobao’s.
Pinduoduo’s product logic is focused on zero decision cost: SKU convergence, concentration on best-sellers, extreme pricing;
Taobao’s advantages are the opposite: higher barriers and value categories like apparel, beauty, home, trendy toys, niche consumption.
Shopping behaviors for these categories are exploratory, driven by chance and excitement—not efficiency or optimal solutions.
Taobao has invested in content, livestreaming, and short video in recent years, all aimed at one purpose: getting users to stay and creating more triggers.
"Browsing" is Taobao’s moat, but AI is naturally the opposite of "browsing".
Of course, Qianwen is trying AI fitting, “planting grass”, and outfit recommendations, attempting to prove value in non-standard categories. But an AI that gets you tissues in 30 seconds and an AI that creates “excitement” in dialogue are several generations apart in product iteration.
For AI shopping, Pinduoduo plays a kind of "villain."
At their March 2026 conference call, Pinduoduo management spent considerable time discussing supply chain and village delivery, but almost nothing on AI.
Silence implies a question: if AI shopping is really a good business, why is the platform most focused on efficiency and low price not in a hurry?
Byte’s Doubao, 345 million MAUs, started internal e-commerce testing with Douyin Mall in March. Doubao can do cross-platform price comparison, even saying some products are cheaper on Tmall and JD than on Douyin. JD launched JD AI Shopping at end-2025, with a minimalist interface (only chat and recommendation). Tencent Yuangbao is integrated with JD, WeChat stores, Dewu; strategy restrained, positioned as decision support, not transaction loop.
By comparison, Alibaba’s moves are indeed the most aggressive.
Qianwen moved from public beta to full integration in half a year; Spring Festival’s 3 billion big free orders drove 130 million people to experience it. Among all players, Alibaba is like the fastest striker in the industry.
But the scenarios where AI shopping is easiest to succeed are not necessarily Taobao’s most profitable ones, which is the core misalignment in Qianwen’s full integration with Taobao.
What’s worrying is that Alibaba invests heavily in pushing AI shopping, but it may actually help the whole industry validate the direction, while benefiting itself less.
After Qianwen’s entry into Taobao, Alibaba’s AI narrative faces new tests.
Previously, Qianwen’s role in boosting e-commerce was mainly imaginative asset.
But once fully integrated into Taobao, the market will inevitably measure with more precise standards: AI shopping conversion rate, Qianwen entry average order value, AI recommendation return rate, incremental GMV or just shuffling existing inventory—these numbers will be closely scrutinized.
4. Entering a Prolonged Battle
For the relationship between AI and e-commerce, Alibaba’s situation is more complex than "doing nothing means losing", but also quite far from "do it and you win".
Taobao’s CMR is growing, users are growing, the basics are stabilizing.
The threat of AI entry to e-commerce is real but far from urgent. The power driving Qianwen’s full integration with Taobao may partly be strategic judgment on AI trends, partly pressure from narrative to capital, partly urgency from competition, partly FOMO memory from the last stall.
These forces intertwined led to a decision of correct direction but somewhat rushed pace.
The inner courage is real; among all Chinese platform companies, Alibaba is the first to aim a large model at its core e-commerce territory, and directly touch the structural contradiction between efficiency and attention.
AI investment is indeed yielding returns: 35.8 billion annualized AI revenue, 40% external cloud growth, BaiLian customers up 8x—the investment is starting to turn into operating cash flow.
But whether AI shopping itself can be a good business remains unproven.
At least for now, the standardized product scenarios it’s best at are not necessarily Taobao’s most profitable; its efficiency may reduce platform dwell time; its inference cost is higher than traditional search; its compatibility with the existing ad system is still unresolved.
Most likely, the outcome will be a middle state: AI shopping becomes a new e-commerce capability, not a replacement; performs strongly in standard products, slowly penetrates non-standard, coexists with the traditional ad system long-term rather than quickly replacing it.
For Alibaba, the entry is defended, but AI shopping seems unlikely to become a disruptive growth engine in the short term—more like a new layer of infrastructure than a new business model.
Taobao’s business model for the past twenty years has relied on users “wasting time”.
The ultimate goal of AI shopping may be to help users “save time”.
How these two coexist within one entity is Alibaba’s core management challenge for the next years.
Just as Disney is still seeking a balance point between streaming and traditional business, Alibaba will walk for a long time on the path balancing efficiency and dwell, convergence and exploration, proof and restraint.
Get it right, it may redefine e-commerce;
Get it wrong, it’s a heavy exploration for the e-commerce industry.
For many things, risk is not only about right or wrong, but arises from anxiety-driven, making something that needs to be done slowly, rushed.
Risk Warning and DisclaimerThe market has risks; investing requires caution. This article does not constitute personal investment advice, nor does it take into account individual users’ specific investment goals, financial situation, or needs. Users should consider whether any opinions, views, or conclusions herein fit their specific situation. Investing accordingly is at your own risk. ```