AMD relations with TSMC are definitely much more tight than Nvidia's. AMD has been helping with developing HPC-optimized nodes and modern packaging techniques

AMD relations with TSMC are definitely much more tight than Nvidia's. AMD has been helping with developing HPC-optimized nodes and modern packaging techniques (3d-stacking and chiplets), while Nvidia tried to force TSMC to drop prices by going to Samsung fabs for 1.5 generations (did not go too well for them). The problem with AMD is definitely not modern node access, but software, and with some investment it can probably change really fast. Apple, on the other hand, is definitely not going to release standalone apple silicon products (GPUs or CPUs), so I don't really believe for Apple Silicon as an AI platform (apart from on-device inferrence).