Chuyên gia Citrini: NVIDIA hưởng lợi từ sự bùng nổ AI mã nguồn mở của Trung Quốc, rủi ro dài hạn nằm ở sự độc lập về tính toán

The Huoxing Finance report states that on July 23, Citrini analyst Jukan published an article refuting the view that "Chinese open-source models are insignificant to NVIDIA," pointing out that the flourishing of Chinese open-source AI is effectively reinforcing NVIDIA's dominant position. China still uses NVIDIA GPUs to train models and releases versions optimized for NVIDIA hardware; thus, for inference, NVIDIA GPUs remain the most advantageous platform for models such as DeepSeek V4 and Kimi K3—no one runs inference for these models on TPUs or Trainium. This protects NVIDIA’s moat in two dimensions: first, by maintaining its status as the optimal platform in the inference market; second, by indirectly dampening the motivation of hyperscale cloud providers to develop their own ASICs to replace NVIDIA GPUs. Meanwhile, Neocloud providers that rely on NVIDIA GPUs for inference also benefit, as the boom in Chinese open-source models provides them with a continuous source of computing demand. Jukan notes that the real structural risk lies in the long term: if China achieves true computing independence and no longer relies on NVIDIA, the consequences will extend far beyond merely losing the Chinese market. NVIDIA’s dominant position would be significantly weakened, the relative competitiveness of alternatives such as TPUs and Trainium would increase, and NVIDIA’s Neocloud ecosystem—considered a key strategic asset—would also suffer.