Ars Technica AI Models & Releases · Jun 24, 2026 22:28 imp:75 OpenAI and Broadcom announce chip designed for LLM inference at scale The silicon race is heating up amid the struggle to keep up with demand. Leer el original en Ars Technica AI →