Q&A with OpenAI VP of Hardware Richard Ho on its Jalapeño inference chip co-designed with Broadcom, using internal OpenAI models to design the chip, and more (Dr. Ian Cutress/More Than Moore)

OpenAI’s Jalapeño inference chip signals that frontier AI is moving deeper into custom silicon to cut inference costs, improve performance, and gain more control over the AI supply chain. For CIOs and technology leaders, this underscores that compute architecture is becoming a strategic advantage—not just a cost center—and IT organizations should expect faster shifts toward specialized accelerators, tighter hardware/software co-design, and more dependence on vendor partnerships for AI capacity.