AVGONew chip, same AI gold rush
Broadcom and OpenAI just announced Jalapeño, an AI accelerator designed around OpenAI’s vision for LLM inference. The pitch is simple: make AI faster, more efficient, and less power-hungry — basically the Silicon Valley version of “do more with less,” but with data centers the size of small cities.
Why investors should care
This isn’t just another shiny press release. Broadcom says the chip was developed from design to production in nine months, with help from OpenAI’s own models. Early testing reportedly shows better performance per watt than current state-of-the-art systems, which is exactly the kind of claim that makes cloud operators start rearranging their capex budgets.
The bigger prize
The companies say Jalapeño is the first chip in a multi-generation compute platform and is meant to be deployed at gigawatt scale with data center partners. Translation: they’re not dabbling. They’re building a full-stack AI infrastructure play, and Broadcom gets a bigger slice of the custom-silicon pie if the rollout sticks.
Big picture
Broadcom already had Wall Street’s attention thanks to its AI networking and custom chip business. This partnership nudges the story further from “nice chip vendor” to “core plumbing for the AI boom.” And in 2026, that’s the kind of narrative investors still happily pay up for.