SpaceX and xAI: Elon Musk assembles super-puzzle from rockets and neural networks

Elon Musk is once again giving lawyers and investors headaches as he reshuffles pieces on his global chessboard. This time, it's about a possible merger between space giant SpaceX and ambitious AI startup xAI. If you thought Grok existed only to ironically respond to users on the social network X, then you clearly underestimated the appetite of the world's richest person. This is not simply an attempt to optimize taxes or simplify management structure, but a strategic maneuver that could change the rules of the game in the race for advanced artificial intelligence.

To understand why this is happening now, we need to recall the context of recent months. Musk is actively redistributing resources among his companies. We've already seen engineers from Tesla transitioning to xAI, and scarce Nvidia H100 chips, originally intended for autopilot, suddenly appearing in data centers of his AI startup.

Musk is building a unified ecosystem where each company acts as a donor for another. SpaceX in this scheme is not just rockets—it's Starlink. A network of satellites enveloping the Earth generates colossal volumes of data and provides connectivity at any point on the planet.

To train next-generation AI, you need not only internet texts, which are already running out, but also real physical data about the world. SpaceX provides it in abundance.

The technical synergy here is obvious to anyone who understands the complexity of modern engineering. Designing Raptor engines or calculating trajectories for Falcon 9 reusable stages—these are tasks where AI can prove itself better than any team of engineers. By combining xAI's brains with SpaceX's manufacturing capabilities and data, Musk creates a closed loop: AI helps build better rockets, and rockets deliver the infrastructure for even more powerful AI. This is a direct challenge to OpenAI and Microsoft, who have no "hardware" in the physical world. While Sam Altman is asking for trillions of dollars to build chip manufacturing plants, Musk already has operating factories and the largest rocket in history, Starship.

The financial side of the question also looks logical, though provocative. SpaceX is valued at over 200 billion dollars and is one of the most stable private companies in the world. xAI, by contrast, is in a stage of aggressive growth and constant search for investment. A merger would allow xAI to lean on SpaceX's giant capitalization, making the company practically invulnerable to competitors in terms of capital attraction. Investors who previously feared investing only in Musk's "software" will now be buying a stake in a company that owns both the sky and intelligence. Of course, this will raise questions from regulators and minority shareholders, but Musk rarely pays attention to such trifles when Mars colonization is at stake.

Ultimately, this merger tells us that Musk has finally become disillusioned with the idea of separate companies for different tasks. He is building what could be called "Musk's universal intelligence." In this vision of the future, Tesla handles vision and movement on land, SpaceX handles expansion and global connectivity, and xAI becomes the central nervous system tying everything together. If Grok gains direct access to Starship telemetry, we could see the first steps toward creating an AI that understands the laws of physics as well as we understand grammar. And this is perhaps the shortest path to AGI that Silicon Valley argues so much about.

The bottom line: Musk is definitively moving toward an "all-in-one" model, turning his companies into a unified technological fist. Will anyone be able to oppose a structure that has both the best neural networks and the only operational heavy-lift rockets in the world?