On May 11, Alibaba announced the full integration of Qwen and Taobao. Users can now select products, compare options, and make purchases on Taobao through conversations with the Qwen app. Taobao has also launched a Qwen-powered shopping assistant.
Before this, global tech giants had spent two years testing the integration of artificial intelligence into shopping. Amazon said its shopping assistant Rufus had accumulated more than 250 million users in 2025 and was expected to bring the company USD 10 billion in additional annual sales. In January 2026, Google also announced partnerships with retailers including Walmart and Shopify to launch AI shopping features in Gemini.
OpenAI, meanwhile, launched ChatGPT’s “Instant Checkout” feature last September, only to announce in March this year that it would abandon the effort.
A global race over whether AI can actually buy things for users has begun, even as companies take diverging paths.
The idea of AI buying things for people has appeared in science fiction for at least 30 years. In Iron Man, Jarvis can complete orders with a few spoken words. In the 2013 film Her, an AI assistant handles everyday tasks through conversation. When Amazon released Echo in 2014, Jeff Bezos envisioned users shopping through AI with a single sentence.
But real-world progress has been much slower than imagined:
- The first phase came during the voice assistant era. From 2014–2018, Amazon’s Alexa Shopping and Google Shopping Actions went online, allowing users to shop by speaking to a smart speaker. But they could handle only simple repeat-purchase commands: a clear category, a clear brand, and a clear quantity. Natural language understanding at the time was limited to basic field extraction. If the request became even slightly more complex, voice assistants could not understand it.
- The second phase was conversational commerce. Around 2018, e-commerce platforms began launching intelligent customer service tools, including Taobao’s Wenwen, JD.com’s Jimi, and Amazon’s customer service bots. But these tools were positioned from the start as after-sales customer service, not shopping advisers. In essence, they were FAQ databases plus finite-state machines, capable of handling returns, exchanges, and logistics inquiries. Asking them to help make shopping decisions was beyond their scope.
- The third phase is the one that has reignited the AI sector in recent years: large models. The bottleneck for earlier AI assistants was ultimately technical. Machines did not understand what people meant. Large models made user intent understandable for the first time.
Starting in 2024, players entered the market one after another. Amazon launched Rufus, Perplexity introduced “Buy with Pro,” and in 2025, OpenAI integrated shopping features into ChatGPT.
The integration of Qwen and Taobao gives AI shopping a more complete form. In practical terms, it has become an agent involved in the core transaction process. It not only understands user needs, but can also call platform capabilities to complete real transactions and services.
Bringing AI into real commercial scenarios has been one of the industry’s central questions over the past two years, as AI moves from answering questions to helping complete tasks. Coding agents, research agents, and customer service agents have already emerged in vertical fields. Shopping is the next scenario everyone can understand: everyone is a consumer, and the transaction loop is naturally closed.
Morgan Stanley noted in a report that agentic spending in US e-commerce is conservatively expected to reach USD 190 billion by 2030, representing a 10% market share. McKinsey, Gartner, and other research institutions have also published market forecasts for 2030. Their specific estimates differ, but they broadly agree on the market’s development trajectory.
The idea that AI will restructure industries has moved from an early prediction at OpenAI’s founding to a mainstream expectation. Yet at the implementation level, large models have evolved more slowly on the consumer side than in office scenarios, because consumer applications involve real products, transactions, and logistics.
Embedding AI into shopping is fundamentally different from using it for writing or coding. It is not only about information processing. It must help users complete a series of real-world actions: buy something, have it delivered, and handle returns if something goes wrong. These steps exceed the capability of a model alone and require a full set of commercial infrastructure.
In 2024 and 2025, global tech giants tried to provide answers. Large model companies were eager to enter real commercial scenarios, while internet transaction platforms wanted to seize the opportunity. They ended up taking three different paths.
- The first path is for model companies to seek external partners, represented by OpenAI. Last September, ChatGPT launched Instant Checkout by connecting with Shopify, allowing users to check out directly in a conversation. But The Information reported in March that the feature was being scaled back and reevaluated after tests showed that users were more inclined to treat ChatGPT as a product research tool than as a transaction endpoint, while too few merchants had integrated with it.
- The second path is for e-commerce platforms to develop their own AI assistants, represented by Amazon. Rufus is embedded directly inside the Amazon app, allowing users to talk with AI, compare prices, and check reviews. According to Fortune, Rufus has accumulated more than 250 million users, and shoppers who use AI are 60% more likely to complete a purchase than ordinary users. But some in the industry have pointed to Rufus’s instability, weaker performance when handling questions outside Amazon’s internal database, and relatively low conversion rate. Part of this has been attributed to Amazon’s proprietary large model, which may not yet match the sector’s leading models.
- The third path is deep integration between large models and e-commerce ecosystems. Alibaba is the representative case. The full integration of Qwen and Taobao is the first deep combination of a leading e-commerce platform and an AI application.
Ultimately, which path a company chooses depends on its resource base.
Model companies lack transaction scenarios and fulfillment capability, so they can only seek external partners. E-commerce giants have complete supply chains and data, but their large model capabilities may be limited. Players that have both large models and real-world commerce have a better chance of building a complete AI shopping experience.
Compared with the US market, China’s longstanding advantage in AI lies in its richer and more mature application scenarios. A huge user base, diverse formats, and rapid market acceptance create complex and varied implementation settings.
China’s major technology companies also have relatively comprehensive application layouts in consumer scenarios. In AI shopping, Alibaba has one of the most complete lifestyle service ecosystems, while Qwen has entered the ranks of internationally competitive large models. With this advantage, Alibaba has chosen to integrate the full AI shopping chain within its own ecosystem, building a barrier that may be difficult to replicate.
Alibaba has more than 20 years of e-commerce capabilities accumulated through Taobao, including search, price comparison, order placement, logistics tracking, and after-sales returns and exchanges. This integration allows those capabilities to be repackaged as “skills” that AI can call.
Inside the Qwen app, users can now complete every step from product recommendation to order placement, fulfillment, and after-sales service, rather than interacting only through shallow redirected links. This is one of the most comprehensive AI shopping service forms at present.
For the model, knowing when to search, when to compare prices, when to place an order directly, and when to advise users to think again requires a large amount of training from real shopping scenarios.
That points to a more important data asset. Taobao’s catalog of about four billion products and more than 20 years of accumulated real shopping scenario data can help Qwen understand purchase intent in user conversations and make recommendations.
Chinese venture capitalist Zhu Xiaohu once said that once the basic capabilities of large models become a relatively stable platform, the essence of competition will shift to engineering implementation capability and building closed data loops. This is precisely where Chinese companies tend to excel.
This reflects a deeper shift in large model competition. As the focus of AI development turns toward value at the application layer, competition is no longer only about model intelligence. It is also about the ability to turn technology into real products, meet specific user needs, and commercialize. As data becomes more important, high-quality data will tend to become closed, turning into each company’s moat.
The significance of this may extend beyond shopping itself. Since January this year, Qwen has been connected successively to service capabilities within Alibaba’s ecosystem, including Taobao Shangou (also known as Taobao Instant Commerce), Fliggy, Amap, and Alipay. The full integration with Taobao further fills a key gap in consumer scenarios.
The same idea is appearing among competitors. ByteDance’s Doubao is accelerating integration with Douyin E-commerce. JD.com and Meituan have each launched independent AI shopping apps, trying to build barriers through vertical scenarios.
Whether users move from testing these tools to using them regularly will take time to answer. But one thing is becoming clearer: if people gradually get used to shopping through conversation, e-commerce, and perhaps the broader consumer market, may no longer look the same.
KrASIA features translated and adapted content that was originally published by 36Kr. This article was written by Xiao Xi for 36Kr.