The switching cost argument is absolutly critical and underappreciated in most NVIDIA coverage. Once training infrastructure and inference pipelines are built on CUDA, rewriting everything for competing architectures becomes a multi-million dollar multi-year project. Google's GEO pivot is similarly smart because they control both the search index and the AI layer, which gives them structural advantages that pure AI startups can't replicate.
The switching cost argument is absolutly critical and underappreciated in most NVIDIA coverage. Once training infrastructure and inference pipelines are built on CUDA, rewriting everything for competing architectures becomes a multi-million dollar multi-year project. Google's GEO pivot is similarly smart because they control both the search index and the AI layer, which gives them structural advantages that pure AI startups can't replicate.
Spot on — it’ll need a catastrophic screw up from Nvidia to allow for any competitor to breath into its market. It’s crazy.