“The semiconductor industry made the AI revolution possible. Now, AI is transforming how chips are built,” said Tim Costa, vice president and general manager of industrial and computational engineering at NVIDIA, as the company announced at SEMICON West that chipmakers, equipment suppliers and EDA companies are adopting its CUDA-X libraries and AI tools to build what it calls AI-defined fabs, connecting accelerated simulation, AI physics, AI agents and digital twins across the factory. The headline product is cuPotential, a new CUDA-X library that accelerates the costliest molecular dynamics calculations; with OpenMM it ran three times faster than an already GPU-optimized baseline. Partners are reporting results across the fab: Lam is training surrogate models that predict plasma behavior in milliseconds, Applied Materials reports up to 35x faster chamber simulations, cuLitho speeds OPC 20x, KLA helped build an image-registration kernel that runs 50x faster than CPU versions, and Advantest’s agents have cut test-cell response times from hours to minutes. Costa said engineers still validate the evidence and decide what reaches production. The next milestone, he said, is validated integration: complete workflows tested on customer data and equipment. Click here to read the full article in the SEMICON West Show Daily. The post NVIDIA Pushes Toward the “AI-Defined Fab” appeared first on Semiconductor Digest .