SpaceX AI: Orbital Compute Dominance Revealed!

[00:00:00]Let's just zoom out. Take a quick snapshot of where we are right now. We already have proof through the Anthropic and Google deals that SpaceX AI is currently capable of doing something with earth-based AI compute cluster that no other company can do. This matters.

[00:00:14]There's nothing stopping a company like Google or Anthropic ordering approximately 12 trillion metric [ __ ] tons of GPUs from Nvidia, building earth-based AI data centers, and then using them for their own needs. Nothing stopping these companies from doing that. Yet they're not. I mean, technically they are, but that's not all they're doing. You cannot ignore the deals with SpaceX AI to access portions of their existing compute. These are astronomical deals. The payback period appears to be a year or two at most, maybe even less. So, riddle me this. If that's the state of things today with earth-based compute, where other companies besides SpaceX AI could be doing yet aren't doing what SpaceX AI is doing with AI clusters, what do you imagine might happen when the enabler for orbital AI compute is Starship, a technology being developed and going along quite well within SpaceX, a company who today are already launching roughly 90% of all of the world's payload to orbit, has already built out a network of roughly 10,000 Starlink satellites. In other words, they're the only company that's going to have the technical capability of putting data centers into space at scale, where how do you think that plays out?