OpenAI says its LLMs helped design the Jalapeño AI chip, cutting development time for an HBM4-based accelerator

OpenAI used its own large language models to help design Jalapeño, an AI accelerator that pairs a compute die with six HBM4 memory stacks and an I/O chiplet. The company says AI assistance drastically shortened the chip’s design timeline and expects future iterations to move even faster.

The project points to a growing trend of AI-assisted chip design, where frontier model developers use their own tools to optimize the hardware they need. If these workflows prove reliable at scale, they could compress semiconductor development cycles and deepen vertical integration among leading AI labs.