I've noticed something important happening in the Chinese artificial intelligence industry these days. Eight years ago, the story of ZTE was a harsh lesson about dependence on foreign technology. But today, the situation is completely different.

The real issue wasn't the chips themselves, but NVIDIA's CUDA platform. This platform has become the foundation of the entire global AI industry. Millions of developers, thousands of applications, a whole ecosystem connected to it. When American export restrictions on chips began in October 2022, Chinese companies felt the true chokehold for the first time.

But instead of giving up, Chinese companies chose a tougher path. They started making radical improvements at the algorithm level. DeepSeek is a clear example: 671 billion parameters, but only 37 billion active during inference. The cost? Just $5.576 million, compared to $78 million for GPT-4. This difference directly impacted prices.

Now comes the biggest step: local chips have moved beyond inference to actual training. Loongson processors and Taichu smart cards are now working on training massive real models. Huawei Ascend has gathered 4 million developers and 3,000 partners. This is an independent ecosystem built from scratch.

The often-overlooked factor: electricity. China produces 2.5 times more electricity than the United States, and industrial electricity in western China costs a quarter to a fifth of the American price. While the US faces a real energy crisis and data centers consume increasing amounts of electricity, China has massive industrial capacity ready.

What’s coming out of China now isn’t products, but (Tokens). DeepSeek is available in 37 languages, supports 89% of the Chinese market, and 40-60% in sanctioned countries. 58% of new AI startups are already using it.

Recent financial reports from local chip companies show revenues growing by 450%, 243%, and 121%, but with significant losses. Don’t misunderstand—these losses aren’t management failures but a tax of war to build an independent ecosystem. Every dollar lost now is an investment in R&D and software support.

The difference from Japan forty years ago is clear: Japan chose to be the best in a global system dominated by others. This time, China is building its own system.

The question now isn’t "Can we survive?" but "At what cost must we pay?" And that cost is progress.