/
Published on 30.01.2026
TLDR: The AI industry is converging with space technology as xAI negotiates a merger with SpaceX ahead of a $1.5 trillion June IPO, while Amazon discusses up to $50B investment in OpenAI. Orbital datacenters with solar power are becoming the next frontier for compute efficiency.
Summary:
There's a fascinating convergence happening that most people haven't fully processed yet. The AI industry and the space industry are merging — not metaphorically, but literally through corporate structures, shared infrastructure, and aligned incentives around energy-efficient compute.
The headline numbers are staggering. xAI is in talks to merge with SpaceX ahead of a June IPO valued at approximately $1.5 trillion. Amazon is discussing up to $50 billion investment in OpenAI, with additional significant sums from Nvidia and Microsoft. The Magnificent Seven of BigTech will essentially have major equity stakes in the three leading AI startups going public: OpenAI, Anthropic, and xAI.
But the strategic logic runs deeper than just capital allocation. SpaceX's groundwork for its massive IPO centers on orbital datacenters as a central pitch to investors. The energy equation is simple: solar arrays in space provide essentially unlimited, cheap power without the constraints of terrestrial energy grids. The engineering barriers are reportedly minor compared to the energy-efficiency benefits. And Blue Origin is actually ahead in manifesting space infrastructure projects at scale.
The speculation gets interesting when you consider OpenAI's potential response. They could acquire Stoke Space or Relativity Space to maintain their compute lead. When Eric Schmidt took over Relativity Space, he understood how key the orbital datacenter race would be with cheap solar power. OpenAI might also acquire 1X Technologies for humanoid robotics — Sam Altman is already a major investor, and the OpenAI Startup Fund led their $100 million Series A.
For architects and teams thinking about long-term infrastructure, this represents a fundamental shift in how we think about compute economics. xAI is burning approximately $1 billion per month. OpenAI's burn rate is comparable. These companies cannot survive on terrestrial infrastructure alone — the energy costs are too high. Grok 5 is currently training on Colossus 2, a supercluster with over 200,000 NVIDIA GPUs, expected to launch when the cluster reaches peak capacity in March 2026.
The Tesla angle is equally compelling. While Tesla's car business has faltered with its first annual revenue drop, its energy business is thriving. Tesla's energy division is highly compatible with the solar arrays needed in space to power orbital datacenters. A Tesla-SpaceX merger before 2028 seems increasingly likely — merging assets is standard practice for Musk, dating back to the 2016 SolarCity acquisition.
Michael Burry's recent comment on Musk is worth noting: "I wrote of incentivized futurists, and there is none more than Elon. I am glad he is here in America. He will go down in history like PT Barnum and Thomas Edison." That comparison — showman and inventor — captures something real about what's happening.
Key takeaways:
Tradeoffs:
Link: Why AI is going to Space
This article was generated from a newsletter summary. The content reflects the source material's perspective and may not represent complete analysis of all aspects of the topic.