NVIDIA 详解 AI 如何重塑半导体晶圆厂工程流程
原标题：AI Is Changing How Semiconductor Fabs Work — and Engineers Are Moving Faster Because of It
AI 摘要
NVIDIA 发布博客，介绍 AI 正在改变半导体晶圆厂的工作方式。通过加速计算、物理 AI、AI 智能体和数字孪生，工程师可以更快完成多物理场仿真、预测等离子体等工艺行为、自动执行工程任务序列，并在虚拟环境中测试产线变更。COMSOL、应用材料、西门子、泛林研究、Advantest 等厂商借助 NVIDIA cuDSS、PhysicsNeMo、cuLitho、NeMo Switchyard 和 Omniverse 等工具，将部分仿真提速 10 至 35 倍，等离子体预测从数天缩短至毫秒级。
正文节选
Demand for advanced chips is rising. The processes to build them are getting more complex. And the margin for error is shrinking. For engineers, understanding how a process change will behave, how a material will interact with a new step, how a defect maps back to a root cause are critical, challenging steps in semiconductor manufacturing — and these analyses can take days. Physical experiments on a fab line are slow and expensive. A wrong call on yield or process quality has real consequences.