Alibaba shares rose 2.6% in premarket trading on Monday after the company moved to unify its artificial intelligence operations under a new business unit led by Chief Executive Officer Eddie Wu.
The restructuring puts Alibaba’s main AI research, product, and enterprise-focused teams under one umbrella as the group seeks faster execution and clearer commercial use for its technology. Alibaba is also due to release quarterly earnings on Thursday, placing added focus on how the company plans to turn AI expansion into revenue.
Alibaba said its newly formed Alibaba Token Hub, or ATH, will focus on building AI work platforms for enterprises. Wu will lead the unit directly. The group will include Tongyi Laboratory, the MaaS Business Line, Qwen, Wukong, and AI Innovation, bringing several major AI functions into a single structure.
The change also ties together more of Alibaba’s AI products and development work. Details tied to the overhaul show the new division will oversee Qwen-related activities as well as AI products linked to DingTalk and Quark-branded devices. The broader scope suggests Alibaba wants closer coordination between model development, software tools, and commercial applications.
Wu described the new group as part of a wider push to coordinate AI strategy across the company while keeping teams flexible enough to move quickly. In his memo to staff, he said the goal is to drive strategic alignment across Alibaba’s AI businesses, deepen internal AI adoption, and preserve speed in execution. This message indicates the company wants tighter operational control as competition in AI intensifies.
The creation of Alibaba Token Hub highlights the company’s stronger focus on enterprise AI customers. Alibaba has already invested heavily in cloud computing and business-facing AI services, and the new structure appears designed to package those capabilities more clearly for companies that want workplace AI tools and model services.
The division’s name also points to monetization. Internal remarks described ATH around the idea of creating, delivering, and applying tokens, a direct reference to the computing units that many AI providers use for billing. This language places revenue generation closer to the center of Alibaba’s AI strategy at a time when many developers in China face pressure to prove they can convert model popularity into profit.
Significant pressure remains across China’s AI market. Chinese firms such as Alibaba, DeepSeek, and Zhipu have offered low-cost or open-source models to expand adoption, but lower pricing has also raised questions about margins and long-term returns. Reuters noted that leading Chinese models can cost 10 to 20 times less than comparable US services, a gap that helps market share growth but complicates profitability.
Alibaba announced the overhaul shortly after the departure of Lin Junyang, the head of its Qwen AI model division. Lin left earlier this month, becoming the third senior Qwen executive to exit this year. The leadership changes drew attention to execution risks inside one of Alibaba’s most important growth areas and increased pressure on management to stabilize its AI roadmap.
Alibaba set up a task force in March to support AI development after Lin’s exit. This group included Wu, Group Chief Technology Officer Wu Zeming, and Alibaba Cloud CTO Zhou Jingren. The formation of Alibaba Token Hub builds on the earlier response and gives Wu even more direct control over the company’s AI direction.
The timing matters for investors. Alibaba plans to post quarterly results on Thursday, and the market is likely to watch for updates on AI spending, enterprise adoption, and product rollout. Monday’s premarket gain suggested that investors viewed the new structure as a sign of tighter focus as Alibaba tries to turn broad AI ambitions into measurable business results.
Also Read: Alibaba Launches Qwen AI-Powered Smart Glasses to Challenge Meta’s Ray-Ban Model