I can't seriously take any statement from Nvidia about the greater good while they maintain an artificial moat of CUDA workloads that can work on other hardware

I can't seriously take any statement from Nvidia about the greater good while they maintain an artificial moat of CUDA workloads that can work on other hardware but is not permitted.