China staged a global artificial intelligence conference in Shanghai on Friday, July 17, where new AI models were unveiled and President Xi Jinping laid out a strategy centered on open, accessible technology, according to Le Monde and bastillepost.com.
Le Monde reports that some Chinese companies are now close behind the leading American firms, and that closing this gap has become central to China's ambitions in the sector. Rather than matching the US on secrecy, Xi is promoting a Chinese alternative that is open and available to the world. Gizmodo frames the pitch bluntly: Xi wants AI to be open to the world and, as it puts it, out of America's control.
Much of the attention has landed on Moonshot AI, which TechCrunch says released a new version of its Kimi model this week. Bloomberg reports that Kimi "upends conventional wisdom" about the US lead over China. Broadband Breakfast describes a Chinese model that took the US by surprise, with abilities rivaling Claude and ChatGPT. TechCrunch notes the release prompted some concern it characterized as fears of "full AI communism."
The common thread across these reports is a shift in tone: where the assumption had been a durable American advantage, Chinese labs are now presenting competitive models and pairing them with an explicitly open, export-friendly philosophy. That combination — capability plus openness — is what Xi is positioning as China's answer to US dominance.
Why it matters: if capable Chinese models are offered openly to developers worldwide, the global AI race becomes less about which single country leads and more about whose approach — closed American systems or open Chinese ones — the rest of the world chooses to build on.