Realtime AI News
OpenAI and Broadcom Unveil Custom AI Chip Designed to Slash Processing Costs by Roughly 50%
OpenAI and chip design giant Broadcom have jointly launched a custom AI chip that aims to reduce AI processing costs by approximately 50%.
OpenAI has partnered with semiconductor giant Broadcom to unveil a custom AI chip designed to reduce AI processing costs by roughly 50%, according to reports. The collaboration marks OpenAI's strategic push into custom hardware development.
The custom chip targets the prohibitively high costs of AI inference and training. By working closely with Broadcom on silicon design, OpenAI aims to reduce its reliance on general-purpose GPUs and gain a significant cost advantage.
The development highlights a broader trend among leading AI companies toward vertical integration, extending from model R&D into chip design, as they grapple with soaring computational demands and cost pressures.
Why it matters
OpenAI and Broadcom's custom AI chip could significantly lower AI deployment costs, potentially reshaping the AI chip landscape and reducing dependence on general-purpose GPUs.