Altman 据称告诉员工，“Eliezer 方案”即一家实验室率先实现递归自我改进并让对手得不到算力，是糟糕的结局，列入只是为了完整。
在这种情形下，某一家开发者率先实现递归自我改进，再凭借失控式的经济优势让所有对手得不到算力——用 Altman 直白的话说，就是“熔掉其他所有人的 GPU”。这是披着思想实验外衣的赢家通吃结局，Altman 据称只是“为了完整起见”才把它列入。原文 (en)
In this scenario, one developer reaches recursive self-improvement first, then uses its runaway economic lead to starve every rival of compute — or, in Altman’s blunt phrasing, “melt everyone else’s GPUs.” It’s a winner-take-all outcome dressed up as a thought experiment, and Altman reportedly included it “only for completeness.”