If you pay attention to the biggest tech companies discussing AI demand, you will notice a common theme: they are running short of compute capacity. This means the large language models that underpin today’s AI products require even more data centers for training and inferencing, and therefore, they need more power. In this context, energy efficiency has suddenly become a critical priority for semiconductor manufacturers.
PowerLattice, a startup founded in 2023 by veteran electrical engineers from Qualcomm, NUVIA, and Intel, claims to have developed a groundbreaking approach that reduces the power needs of computer chips by more than fifty percent. On Monday, the startup emerged from stealth with a twenty-five million dollar Series A funding round led by Playground Global and Celesta Capital, bringing its total funding to thirty-one million dollars.
Pat Gelsinger, general partner at Playground Global, stated that this work addresses the difficult challenge of how to get power into a device, noting that very few teams and people can do it. He described the PowerLattice team as the dream team of power delivery. As the former CEO of Intel, Gelsinger carries significant authority in the semiconductor world, making his participation a powerful stamp of approval for the startup.
In fact, Gelsinger shared that when the startup’s CEO, Dr. Peng Zou, and the founding team pitched their idea at Playground’s offices in March, they were so star-struck by his fame that they asked for a selfie. The admiration proved mutual, as Gelsinger came away genuinely impressed with PowerLattice’s technology.
The startup’s technology is simple in concept: a tiny power delivery chiplet designed to bring power closer to the processor, which significantly minimizes energy loss. Two years in, PowerLattice has already achieved its first key milestone. Its first batch of chiplets is being produced by TSMC, in partnership with an unnamed manufacturer that is testing the startup’s functionality.
Beyond its initial customer, the startup plans to make its product available for testing by other customers in the first half of 2026. These trials should prove instructive, given that PowerLattice’s potential customer set includes major chip manufacturers like Nvidia, Broadcom, and AMD, as well as specialized AI chip developers such as Cerberus, Grok, and Playground-backed startups d-Matrix and NextSilicon.
While every chip company has internal teams working on improving energy efficiency, Gelsinger hopes that PowerLattice’s innovative approach will pique their interest. He suggested that customers might allocate some volume to this new approach and some to their more traditional methods, but he believes PowerLattice’s ability to capture meaningful market share will quickly emerge.
PowerLattice is not the only startup attempting to help chip manufacturers address the energy problem. The company would compete most closely with Empower Semiconductor, a startup that raised a one hundred forty million dollar Series D round led by Fidelity Management and Research Company in September.
However, Gelsinger is certain that PowerLattice’s fifty percent energy efficiency gain is an extraordinary result, and he expects the company to soon raise a much larger funding round to fund production. He said the idea is bold, the benefits are large, and he expects others will be saying it is a great idea and trying to emulate it.