One GPU per prompt — 7 billion AI users implies staggering GPU demand
"DACM Insights: Decentralizing AI, The Akash Approach" (DACM Insights)
“Inference — in order to infer a model, you need one GPU per one prompt. Simply put, every prompt you send… takes 10 seconds to return the answer; that 10 seconds is being computed on one GPU… So if you have 7 billion people in the world, if everybody’s on AI, you can imagine how many GPUs we need.” — 00:13:55
Context: Answering why AI needs so much compute/energy, as setup for the decentralization argument.