Samsung reported that first-quarter profit likely surged more than 700% year-over-year, driven by explosive demand for high-bandwidth memory chips used in AI training and inference systems. The number is staggering but follows a period when Samsung’s memory business was being undercut by inventory gluts and pricing pressure.

The reversal is a clean signal: the AI buildout has moved deep enough into the stack that it is now pulling on memory, not just logic chips. Nvidia gets most of the narrative oxygen when people talk about AI hardware, but the demand chain runs straight through DRAM and HBM suppliers. Samsung, SK Hynix, and Micron are all beneficiaries of the same underlying dynamic.

Everyone knows Nvidia makes the chips that power AI. Fewer people understand why competitors have been unable to close the gap despite years of trying and billions of dollars of investment.

The hardware advantage is real but secondary. AMD makes competitive GPUs. Google has TPUs. Amazon and Microsoft have custom silicon. On raw performance for certain workloads, these alternatives are credible. The reason Nvidia keeps winning is not the chip — it is CUDA.