Filtered to AI Hardware & Semiconductors · show all articles
Nvidia versus AMD gets the headlines. The rivalry that decides which hyperscalers can build competitive custom silicon at all runs between two quieter companies most readers couldn't name.
Google's TPU, and at least five other hyperscalers' custom accelerators, share a co-designer almost nobody outside the industry can name.
OpenAI's custom-accelerator agreement with Broadcom is larger than any other single deal in this cohort by power commitment. This briefing works through the numbers, the timeline, and what building at this scale actually requires.
Not the chip. Not the training cluster. Ayar Labs focuses on one narrow, unglamorous problem — getting light on and off a chip efficiently — and is betting that narrowness is exactly what makes the bet work.
Almost every company in this cohort's accelerator and startup tracks is American, Korean, or Chinese. Europe's near-total absence from that list is itself a story — and Axelera AI is one of the few companies trying to close the gap.
No EUV lithography machine, no leading-edge chip — full stop. Every AI accelerator in this cohort exists because exactly one company in the world can build the tool that makes it possible.
ASML's next-generation lithography platform isn't a minor refinement. It's the specific technology that determines whether transistors keep shrinking past the current generation at all.
A finished wafer isn't a sellable product. Someone still has to cut it, package it, and test it — and the largest company doing that job, ASE Technology, holds nearly half the entire advanced-packaging market.
ASML gets the attention because it has no competitor. Applied Materials' job is just as essential and far less understood: depositing the precise atomic layers every chip is actually built from.
Every other company in this cohort sells silicon, licenses it, or manufactures it for someone else. Apple designs some of the most advanced chips in the industry and has built its entire strategy around never putting one up for sale.