On August 7, according to a report by The Information, NVIDIA is considering a significant move to address the shortage of high-end high-bandwidth memory chips by planning to adopt a lower memory configuration than originally planned for its next-generation graphics processor, Rubin Ultra. Sources familiar with the matter said that NVIDIA has tested at least three versions of the Rubin Ultra graphics card over the past few weeks, with some versions featuring a lower memory capacity than the initial specifications. The adjustment is due to the potential inability to secure sufficient high-end memory to meet the supply demands of the original design. While reducing memory will affect chip performance, NVIDIA can compensate through other means. However, other AI companies using this reduced-memory chip to run large AI models will need to deploy more chips.