The lab’s chief applied scientist attributed the closure to ’strategic adjustments between China and the United States.’
Amazon is closing an artificial intelligence (AI) lab it established in Shanghai seven years ago, the latest American company to scale back its China-based research force amid ongoing U.S.–China tensions.
The lab, opened during the 2018 World Artificial Intelligence Conference (WAIC), was Amazon’s flagship investment in overseas AI research. News of its closure comes just days before this year’s WAIC is set to take place in Shanghai.
The decision to shut down the Shanghai lab was first publicized by Wang Minjie, the lab’s chief applied scientist, in a statement posted Tuesday on the Chinese messaging platform WeChat. He attributed the closure to “strategic adjustments between China and the United States.”
Wang said his team had published more than 100 research papers and built from scratch the Deep Graph Library (DGL), an open-source framework for machine learning on graph-structured data. According to Wang, DGL has generated nearly $1 billion in revenue for Amazon’s e-commerce business.
“I can confidently say that my team ranks among the very best in the field of Agentic AI, in terms of technical depth, scientific rigor, and execution,” Wang wrote, referring to a class of AI systems capable of setting their own goals and taking actions with minimal human intervention, in contrast to traditional AI that primarily responds to user input.
Amazon confirmed the lab’s closure following Wang’s announcement. The company did not indicate whether the current geopolitical climate had to do with the decision, but emphasized that it was the result of an internal strategic review.
“After a thorough review of our organization, our priorities, and what we need to focus on going forward, we’ve made the difficult business decision to eliminate some roles across particular teams in AWS,” Amazon spokesperson Brad Glasser said in an emailed statement, referring to Amazon Web Services, the company’s cloud computing service.
AWS, which offers a suite of AI and machine learning tools, remains Amazon’s most profitable division. However, in its most recent quarterly earnings released in May, AWS missed revenue expectations for the third consecutive quarter. Despite a 17 percent year-over-year increase to $29.3 billion, growth slowed from 18.9 percent in the previous quarter, marking the lowest growth rate in more than a year.
“We didn’t make these decisions lightly, and we’re committed to supporting employees throughout their transition,” Glasser said.
