aifollow.news
OpenAI CEO Sam Altman argues that, amid tight DRAM supply, concentrating memory in higher utilization data centers could better meet overall computing demand than leaving large amounts in personal computers. Samsung Electronics expects supply to remain tight through 2028, with new capacity unlikely to raise supply substantially in the near term.
Coverage of the same occurrence shares a node; subsequent developments have their own nodes.
Sam Altman argues that, from a global resource allocation perspective, it is inefficient for limited DRAM to sit in personal computers and be used only 5% of the time. The post’s author says datacenters are more efficient per token, while local models cannot be deprecated by a service provider.
By original publication time, with each report's bookmarks and feedback preserved.
OpenAI CEO Sam Altman argues that, amid tight DRAM supply, concentrating memory in higher utilization data centers could better meet overall computing demand than leaving large amounts in personal computers. Samsung Electronics expects supply to remain tight through 2028, with new capacity unlikely to raise supply substantially in the near term.
Asked about RAM shortages, OpenAI CEO Sam Altman said that, from a global allocation perspective, it is “quite terrible” to have people run AI models locally on their own computers. He was responding to the view that open-source models running on personal PCs would remove the need for data center resources.
Sam Altman argues that, from a global resource allocation perspective, it is inefficient for limited DRAM to sit in personal computers and be used only 5% of the time. The post’s author says datacenters are more efficient per token, while local models cannot be deprecated by a service provider.
Name a person, company or topic. We find the sources and gather every new development in your own tab.