Alibaba's Qwen AI chatbot halts coupon issuance after user surge

Alibaba’s AI chatbot Qwen temporarily halted issuing coupons due to high user demand, impacting a promotional campaign during China’s Spring Festival.

The initiative, part of a US$430 million plan, aimed to encourage in-app purchases through chatbot prompts.

Since launching last week, Qwen has aimed to integrate Alibaba’s apps and enable direct payments, similar to features offered by other tech companies.

Alibaba reported that 10 million orders were placed within the first nine hours of the campaign.

Over the weekend, the chatbot experienced technical overload, leading to a message on its official Weibo channel requesting users to be patient.

On February 9, users received refusal messages when attempting repeat purchases, citing excessive demand.

🔗 Source: Reuters

🧠 Food for thought

Implications, context, and why it matters.

Qwen positions Alibaba toward an all-in-one AI assistant with agent-like skills

Dense technical specs make Qwen more than another chatbot. It sits at the center of Alibaba’s “agentic AI strategy” for real-world task execution 1.
Qwen works as a “task-completing agent” that pulls core services into one chat flow, including Taobao (Alibaba’s main e-commerce marketplace), Alipay (Alibaba-affiliated payments app), and travel-booking platform Fliggy 2.
A single voice or text prompt can trigger multi-step actions such as ordering food, booking travel, or paying for a purchase 3.
Alibaba wants one assistant across its ecosystem, while also trying to make money from agentic AI faster than rivals like ByteDance’s Doubao 4.

AI agents could squeeze search and marketplace ad models

Alibaba reports 10 million orders in nine hours, a result that implies shoppers may adopt AI agents for commerce at scale 5.
That momentum can clash with search and marketplace ads that fuel Taobao and also support Google’s search business.
When an AI agent picks the best product for a user, merchants may gain less from paying for top placement, which can “compress” today’s ad model 6.
This change could push the industry from selling attention through ads to earning revenue from completed tasks, a hard shift for companies built on search traffic 6.