OpenAI Unveils Jalapeno AI Inference Chip Built with Broadcom - Tech | Business | Economy

▤ Summary

OpenAI has introduced Jalapeño, its first custom AI inference chip, built with Broadcom to reduce dependence on external chip suppliers. The chip targets inference workloads, where AI systems answer user requests at scale and consume significant compute and energy. The article explains why specialized inference hardware can lower operating costs, improve efficiency, and speed up responses in data centers. It also notes that the trend is part of a broader move among major AI companies toward in-house silicon. The piece closes by linking inference chips to the future of AI services, while pointing out that consumer devices already use separate on-device AI accelerators.

▦ Linked Companies

Broadcom Inc.

$50M to $100M

OpenAI

$25M to $50M