During an appearance on the Cheeky Pint podcast, OpenAI CEO Sam Altman was asked about supply chain shortages, including the ongoing RAM shortage. He said it is “terrible” from a global allocation standpoint to have RAM distributed across personal PCs rather than concentrated in data centers. Altman also pushed back on the assumption that open-source models, run locally on users’ own computers, would reduce the need for data center infrastructure. He noted that many people want local inference but argued this view overlooks allocation concerns. He also commented on AI’s impact on math during the same interview.
Why it matters
The remarks touch on an active debate over whether AI inference should run locally on consumer hardware or remain centralized in data centers, a question tied to the current RAM shortage and the rise of open-source models.
Who should care
Readers following AI infrastructure, hardware supply chains, and the trajectory of local versus cloud-based inference.