Summary

• The objection to domain-specific silicon, which might sound quite persuasive, is the one about obsolescence.
• A chip may take two to three years to design, fabricate, and deploy, and AI moves fast enough that whatever you tuned it for has usually moved on by the time it lands in a data center.
• Google’s bet on TPU was somewhat contrarian, but thankfully, that bet has been on the right side of history so far.
• The latest ASIC that grabbed a lot of attention is OpenAI’s Jalapeño chip, its first in-house inference chip, built with Broadcom and Celestica.
• Such fast-paced, LLM-optimized chip design does raise questions around the depth of the moat around Nvidia’s CUDA.

My friend Liberty recently highlighted an interesting paper from Google (GOOG) (GOOGL) that goes through the evolution of TPUs. The objection to domain-specific silicon, which might sound quite persuasive, is the one