SpaceX needs money badly. The merger was the first step to solve it as so far xAI is very disappointing (performance wise, Grok isn’t a good model like Opus or ChatGPT1). Burns cash like hell with no end in sight. The S-1 makes it clear that the biggest future revenue will come from AI, the microhard and Terafab project. It also states that both are ideas which are currently in a legally non-binding state, so they might not even happening. Intel can quit anytime, it’s not something I’d bet my money on.
Lastly, the whole data centers in space is still something that most likely won’t work. They would need a huge surface area away from the sun to dissipate the heat generated by the AI chips. Not going to work IMO.
I strongly advise anyone interested to do a lot of research into the feasibility of these projects as currently only Starlink makes money and their TAM is limited (they also now get a lot of competition from Bezos and European companies, China as well).