US AI restrictions could give Chinese competitors an edge
Recent restrictive measures imposed by the White House on leading American AI developers may inadvertently hand a strategic advantage to Chinese competitors, according to CNBC. Following an export restriction, Anthropic was forced to halt its activities for two weeks, after which it ultimately received limited permission to release its Mythos 5 model to specific federal agencies and companies, while its Fable 5 is still not available. OpenAI has also scaled back the introduction of its GPT 5.6 models in response to requests from the government.
Market leaders and government officials have previously argued that minimising regulatory barriers was essential for the US to maintain its lead. Critics, however, have suggested that the current restrictions are hampering domestic progress just as China is closing the gap.
For example, Zhipu’s GLM 5.2 model is reportedly on a par with leading US labs in certain cybersecurity benchmarks. Venture capitalist Marc Andreessen pointed to the unfortunate timing of these developments, while analyst Christopher Wood stressed that GLM 5.2 offers comparable business advantages at a fraction of the cost.
This shift coincides with a transition in the corporate world from unlimited AI spending to a demand for greater efficiency and better returns on investment. Economic pressure is pushing American companies towards Chinese alternatives. The CEO of AI start‑up Lindy has already recently migrated all traffic to DeepSeek in order to drastically reduce costs.
Because many Chinese models are ‘open‑weight’, companies can easily download and host them locally, bypassing the traditional dependence on the cloud. Thanks to this flexibility, companies such as Coinbase and Shopify have been able to deploy models like Qwen 3, Kimi 2.7 and GLM 5.2 to cut costs while simultaneously expanding functionality.
The situation is leading to a complex geopolitical paradox. While the US has strictly limited China’s access to high‑end hardware such as Nvidia and AMD chips, the gap in software is steadily narrowing.
Despite the hardware bans, the leadership of Zhipu and Elon Musk have suggested that Chinese models will soon match the performance level of the most advanced American systems.
Experts are particularly concerned about the implications for national security and cybersecurity. Some specialists warn that open‑weight Chinese models are becoming increasingly adept at automating complex cyberattacks.
There is a growing fear that by restricting domestic AI development, the US government is hampering its own industry in building the necessary defensive mechanisms to cope with the evolving capabilities of Chinese AI.
Business AM
Subscribers 0
Fans 0
Followers 0
Followers