Sources: Alibaba names senior AI researcher Dayiheng Liu as the head of its Qwen LLM project, bringing clarity after multiple rounds of reorganization in 2026
Alibaba Group has appointed Dayiheng Liu, one of its senior AI researchers, as the head of its Qwen large language models …
Context & Ripple Effects
Qwen’s leadership has been unsettled since Junyang Lin’s abrupt departure in March, which was followed by Alibaba’s foundation-model task force. In April, Alibaba put Cloud CTO Zhou Jingren over the wider AI unit amid a reported move toward monetizable model-as-a-service offerings.
The reported choice of Dayiheng Liu would give the model program a named technical leader after that reorganization cycle. It also follows a burst of proprietary Qwen releases, including three closed-source models launched within three days, while Lin’s new lab has emerged as a separately funded AI contender.
First-order effects
- If confirmed, Liu becomes the accountable technical lead for Qwen, clarifying decision-making for the researchers and product teams reshuffled under Alibaba’s broader AI leadership.
- Alibaba Cloud enterprise clients using Qwen products gain a clearer owner for the model roadmap, including work adjacent to Tongyi Lab’s robot-model pilot.
Second-order effects
- Alibaba’s effort to commercialize Qwen faces a sharper talent-retention test: Lin’s well-funded new lab is competing for attention from the same pool of foundation-model researchers.
- A stable Qwen lead would help Alibaba align rapid proprietary-model releases with its MaaS strategy, making model quality and enterprise adoption more tightly connected.
Third-order effects
- If Alibaba sustains this leadership structure, China’s leading model programs are likely to be organized more directly around cloud distribution and enterprise revenue rather than standalone research teams.
- The split between Alibaba’s commercial Qwen effort and former Qwen leaders’ independent labs points toward a more competitive domestic market for model talent and funding.
The trend: Foundation-model groups are being folded more tightly into cloud businesses as AI companies seek leadership stability and revenue paths alongside research progress.