Two things in AI caught my attention today.
首先, 主权 AI , when our data, prompts, 工作流 and actions 通过 through external model providers, the question is no longer just which model is the smartest. It is also about who controls the data, infrastructure and intelligence behind our operations.
Second, model distillation , Chinese AI giants are showing that smaller, specialised models can be distilled from frontier models and still achieve highly competitive capabilities against leading US models.
These two developments make me even more confident that we are on the right path.
We are building our own 智能体 AI architecture, while also developing distilled LLMs using our own operational data and 中小企业 use cases.
The goal is not simply to chase the biggest model.
It is to build AI that is more sovereign, specialised, efficient and practical for real 中小企业 operations.
That is the direction we believe in.
#主权AI #AgenticAI #LLM #模型蒸馏 #AI基础设施 #中小企业 #AINNA #Artificial智能
Ruang pembaca
Apa pendapat anda?
Komen disemak sebelum diterbitkan. 电子邮件 diperlukan untuk moderasi dan tidak dipaparkan kepada pembaca.
这篇文章对infrastructure and intelligence behind our的解释很清楚，实际操作的重点也很容易理解。
关于infrastructure and intelligence behind our的实际落地部分最吸引我。
我喜欢文章对infrastructure and intelligence behind our保持务实的态度。
文章对infrastructure and intelligence behind our的结论比较平衡，不只是强调好处。
视觉和结构让infrastructure and intelligence behind our的概念更容易掌握。
关于这篇文章这部分写得很扎实，整体也站得住。