(Reuters) -Chinese search engine company Baidu on Wednesday reported a drop in second quarter revenue as its core advertising business struggled amid China's economic slowdown, while returns from AI investments remained limited.

About 500 robot athletes from 16 countries competed in Beijing as the United States and China race each other to shape the future of AI.

Sam Altman has cautioned against the use of export controls and has warned that China can probably build inference capacity faster than the U.S..

With the plan finalized earlier this year and modest compared with U.S. hyperscalers' AI capex trajectories, it underscores Alibaba's efficiency-first approach to AI R&D while also reducing its vulnerability to near-term disruptions from Beijing's scrutiny over the use of Nvidia's H20 GPUs.

In a surprising reversal of the United States’ years-long technology restrictions on China, President Donald Trump last month allowed Nvidia to resume sales of a key AI chip designed specifically for the Chinese market.

Behind closed doors, Chinese researchers are laying the groundwork for a new global AI agenda—without input from the US.

I’m worried about China,” the 40-year-old Altman was quoted as saying on Monday in a report by American business news channel CNBC. He indicated that the AI arms race between the US and China was more complex than it appeared.

As Washington tries to limit its progress, Beijing is spending more to build an artificial-intelligence ecosystem that doesn’t rely on U.S. technology.

Despite U.S. export controls on advanced semiconductors, the company had managed to develop a customizable open technology that could compete with some of the most advanced proprietary American AI models,

China is “set up to hit grand slams,” longtime Chinese energy expert David Fishman told Fortune. “The U.S., at best, can get on base.”

U.S. companies and policymakers are mobilizing their response to free-to-use AI models from China.

But as their new plans illustrate, that competition may also divide the world into competing realms of AI products and governance.