研究显示中国最强 AI 模型训练使用了被禁运的 Nvidia 芯片
Export-banned Nvidia chips are behind China’s most powerful AI models—uh oh, Jensen!
这条值得点开看看：研究实锤了 DeepSeek、Qwen 这些中国头部模型其实是用被禁的 Nvidia 芯片训练的。对关心出口管制走向、或者做合规判断的人来说，这会影响后续政策收紧的预期。
最新研究指出，中国最先进的 AI 模型在训练过程中使用了美国明令禁止出口的 Nvidia 芯片。该研究通过硬件指纹分析等方法追踪模型训练所用的芯片型号，结果显示被禁的高端 GPU 仍然流入了中国的头部模型训练管线。
这一发现直接指向美国出口管制政策的有效性问题：管制本意是限制中国在高端 AI 算力上的获取能力，但被禁芯片仍出现在 DeepSeek、Qwen、Kimi 等头部模型的训练环节中。对 Nvidia 而言，这既带来合规审查压力，也可能引发更严格的执法；对中国实验室来说，算力获取渠道的合规风险将持续存在。
Export-banned Nvidia chips are behind China’s most powerful AI models—uh oh, Jensen!
New research shows the most advanced AI models from China were trained on banned Nvidia chips.
This report is categorized under AITOP's "Industry" desk, focusing on China's AI industry and domestic foundation model ecosystem. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison and AI regulation, compliance frameworks, and sovereign compute policy. Key entities and technical topics under continuous tracking include Nvidia, DeepSeek, Qwen, Kimi.
Source attribution: curated from techcrunch (@Tim Fernholz). Cross-source corroboration: 11 independent sources have covered this development (including rohanpaul_ai, The Business Times: Tech, VnExpress Số hóa), confirming industry-wide signal strength.
