Reading up on ai-hardware

Control over semiconductor supply chains, from ASML's lithography to TSMC's manufacturing and NVIDIA's CUDA ecosystem, determines which entities dominate the AI era.

SpaceX plans a $55 billion investment in a new semiconductor factory, Terafab, to manufacture AI chips and deepen Elon Musk's push into artificial intelligence.

Musk's TERAFAB plan bypasses ASML's EUV bottleneck by using Intel GaN chiplets on mature nodes and advanced 3D packaging to achieve 1 TWh compute capacity.

Qualcomm's Ventuno Q single-board computer combines a Dragonwing IQ8 processor with a microcontroller and 16GB RAM for offline edge AI and robotics applications.

SRAM-centric chips outperform GPUs on memory-bound AI inference workloads, particularly decode, by placing memory near compute for faster bandwidth.