China deals a strong blow to American artificial intelligence dominance

China’s leading AI companies are ramping up pressure on Silicon Valley, with Moonshot and Alibaba unveiling models they claim can compete with the best of what OpenAI and Anthropic have to offer at a fraction of the cost. Rapid releases suggest that America’s advances in artificial intelligence are becoming increasingly aggressive, just as the technology is becoming central to national security, economic power, and geopolitical influence.

The opening salvo came from Beijing-based Moonshot AI, one of China’s leading AI model developers, which Kimi K3 unveiled Friday. Moonshot Claims Its own tests consistently rank it above almost all US systems, lagging only OpenAI’s GPT-5.6 Sol and Anthropic’s Cloud Fable 5, though it is ahead in some benchmarks.

Over the weekend, Chinese tech giant Alibaba followed suit with Preview From Qwen3.8, a new model that it says is “one of the most powerful models available today” and “second only to Fable 5,” Anthropic’s flagship model.

The two companies emphasize a key difference from leading American laboratories: Instead of locking their most advanced models behind closed doors, they make them available to the public. While some US companies, most notably Meta, have taken a similar approach, releasing models that developers can freely download, modify and build upon has become a growing point of differentiation for China’s AI industry.

Moonshot Kimi describes K3 as the world’s largest open source AI system, with 2.8 trillion parameters. The number of parameters are measures of the model’s complexity during training and provide a rough indication of its size and performance, although larger does not always mean better. Alibaba says Qwen3.8 is a 2.4 trillion-parameter model that is “continuously evolving.” Neither OpenAI nor Anthropic disclose the exact parameter count for their flagship systems.

It remains difficult to assess the capability of either Chinese model until they are fully launched and tested independently. Moonshot says it will release the full model weights — the internal numerical numbers learned during the AI ​​model’s training period — a week from now on July 27. Alibaba says Qwen3.8 “will be open-weighted soon.”

However, these releases have already intensified the competition between the United States and China in what has been repeatedly described as the defining technology race of our time. It has shaken up the industry in a way not seen since DeepSeek unveiled a low-cost model last year that rivals leading American systems. The models also raise questions about whether the vast amounts of money American companies spend on chips, data centers, and model training can guarantee a lasting advantage, especially if Chinese rivals are able to approach — or exceed — those limits with fewer resources.

The prospect of releasing two highly capable Chinese models for others to download and adapt also stands in stark contrast to the more guarded approach taken by American labs, whose most advanced systems remain proprietary. This openness is evident even as Washington moves quickly to restrict global access to key technology. The government has used export controls to restrict China’s access to the most advanced chips, as well Forcing Anthropic to withdraw its most capable system from the market due to concerns that it may help foreign competitors catch up.

It remains to be seen whether the new Chinese models live up to their creator’s claims. But, as with DeepSeek before them, this technology is likely to intensify the technological rivalry between the United States and China, influence economic and national security policy, and show that America’s lead is much narrower than it once seemed.