Quote · Moonshots with Peter Diamandis
SpaceX IPOs at $2.89T Market Cap, US Govt Suspends Fable & Mythos 5, Altman Delays OpenAI’s IPO | EP #265
Where this was said
Dave and Salim on Altman's Strategic Position and Intelligence as Utility
At 1:02:04 · chapter starts 59:05
Dave argues Altman has no personal financial stake in rushing the IPO and may be struggling for relevance. Salim proposes Salim's Law: every 10x token drop yields 100x more experiments. [1] — Salim Ismail "Every 10x drop in tokens means we're doing 100 times more experiments." 1:02:04
Sam Altman said the faster recursive self-improvement takes off, the more it makes sense to delay OpenAI's IPO. The shallow read: Codex revenue isn't ready. The deeper read: if a company can improve itself without needing external capital, this is the earliest signal of technology decoupling from capital — a preview of what a post-capitalist economy looks like.
Salim Ismail proposed that every 10x drop in the cost of AI tokens enables 100 times more experiments, compounding AI's impact on innovation.
OpenAI's Codex can now set its own goals. Developers on the team no longer write their own objectives — they ask Codex to write a goal for itself and for each sub-agent it spawns. This is the inflection point where you stop telling AI what to do and start telling it what you want.
