I just heard a piece of news that has been stirring the AI industry circles lately. Lin Junyan, the chief technologist who elevated Qwen to the top of the globally open-source large language models, unexpectedly resigned from Alibaba in March.

The story started dramatically. On March 3rd, one day after a major strategic meeting focused on AI, Lin Junyan posted a short tweet: "me stepping down. bye my beloved qwen." The next morning, he officially announced his resignation. What drew more attention was a comment from one of the key contributors to the team who said: "I am truly heartbroken, I know the departure wasn’t your choice." This strongly indicated that the resignation was not voluntary.

But why did this happen? Information points to deep disagreements within Alibaba regarding the development direction of the models. There was a plan to radically restructure the Qwen team, dividing the integrated team into separate units. Lin Junyan believed this would weaken competitiveness, as the strength of developing large models comes from deep collaboration across all stages.

Deeper than the technical disagreement, there was a strategic conflict. Under Lin Junyan’s leadership at Alibaba, Qwen adopted a comprehensive open-source strategy and rose to global prominence. But senior management at Alibaba began to focus more on commercial goals and revenue rather than building technical standards. Some managers described the recently launched Qwen-3.5 model as a "unfinished product" that hadn’t been refined yet, reflecting dissatisfaction with its commercial performance.

The impact is now clear. Within a few weeks, several key leaders of Qwen have left: the subsequent training manager, and the main contributor to the code. This threatens the pace of future model development. Alibaba is now trying to persuade Lin Junyan to stay, but it’s likely he will start his own project or join another team in the field.

What’s more interesting is what this means for the broader scene. Lin Junyan, born in 1993, is considered one of the rare local tech leaders in China who grew entirely within the industry without needing abroad experience. He studied computer science at Peking University and specialized in linguistics. After only six years at Alibaba, he reached P10 level at age 32, the youngest person to do so in the company. This reflects his exceptional importance.

His departure marks a real turning point. Alibaba is shifting from building technical standards and an open ecosystem to focusing on commercial transformation. But the cost could be high: losing core team members, declining morale, and increased competition from ByteDance, Tencent, and others. These organizational tremors will truly test Alibaba’s strategic resilience in AI.