Tesla Selling Shanghai Gigafactory for SpaceX Merger - Elon Musk Says It's Fake News

Eli opens by acknowledging that the premise of his own episode sounds absurd on its face: Tesla reportedly weighing the sale of its China business to smooth a potential merger with SpaceX. He frames his entire approach around a joke that has become his working rule for 2026, that anything sufficiently ridiculous is probably true. The Wall Street Journal, citing a source familiar with the talks, reported that Tesla executives have been told to prepare for a separation of the China business ahead of a possible SpaceX tie-up, with advisers reportedly discussing a spin-off, sale, or outright closure. Crucially, nobody at either company has flatly denied the merger talk; SpaceX president Gwen Shotwell told CNBC in June that combining the firms might make Musk's life easier by streamlining management, a rationale Eli finds telling precisely because it centers one individual's convenience rather than shareholder value in two trillion-dollar public companies.

Much of the episode's energy comes from Eli walking through SpaceX's brutal post-IPO trajectory. Shares priced at $135, opened around $150, spiked to $225 and briefly valued the company near $2.9 trillion within a week, then crashed to about $109, with the low touching $107. A lockup expiration roughly two weeks out will triple the number of shares available on the open market, which Eli predicts will push the price down further. He argues the valuation never made sense in the first place, contrasting it with Tesla's story: he disagrees with Tesla's humanoid-robot-driven trillion-dollar bull case but at least understands the argument, whereas with SpaceX he says he cannot even locate a coherent thesis.

To puncture the mythology around SpaceX's launch business, Eli reaches for two deliberately absurd comparisons. He notes that SpaceX's 2025 space launch revenue was about $4.1 billion, barely ahead of GameStop's $3.6 billion in yearly revenue, the self-styled grandfather of meme stocks. He revives an older comparison too, that SpaceX's launch revenue sits in the same range as OnlyFans, a point he uses not as a moral judgment but as a blunt measure of what the market actually pays for space access versus its mythology. He adds that Starlink is just one ISP among many, and that SpaceX's touted AI1 space-based data centers amount to a single server rack per satellite, which he calls dumb on its face rather than a serious computing platform.

Eli traces what he sees as a repeating pattern in Musk's recent corporate history: Twitter, purchased and then run into the ground, losing roughly three quarters of its value from about $44 billion down to near $12 billion, was absorbed into X AI under the justification that AI needed to train on real human communication. X AI then became a cash-burning product nobody wants, illustrated by his claim that federal employees with free access to Grok still route around it toward Anthropic, OpenAI, and Google. He argues that X AI's troubles then got folded into SpaceX, and that the same absorption logic now points toward folding a struggling SpaceX into Tesla, since Tesla still carries brand strength and public confidence that SpaceX currently lacks.

The China angle is where Eli grounds the story in real regulatory stakes. Gigafactory Shanghai is Tesla's largest and most productive plant, historically producing more than half of global deliveries with an annual capacity above 950,000 vehicles, making it look like a crown jewel no rational company would shed. But Eli points out that SpaceX handles sensitive US government work, including spy satellites and the Star Shield military ISP program used for drone control operations referenced in Ukraine and against Iran, which JP Morgan analysts flagged as a practical bottleneck to any merger because of national security concerns tied to SpaceX's government relationships operating inside China. He also raises a separate, more defensive rationale: even without a merger, Tesla executives may want a clean way to separate the China operations simply to protect against risks like nationalization, especially if Chinese and US operations share components, employees, or intellectual property that would make a forced seizure catastrophic rather than merely costly.

Eli revisits Musk's own past statements that Tesla's vehicles are essentially worthless without solved full autonomy, and argues Musk is now steering the company toward Optimus humanoid robots as its real long-term identity. If that pivot is genuine, a factory built almost exclusively to manufacture cars, however productive today, becomes a less strategic asset for tomorrow, which Eli says helps explain why Tesla might be willing to hand Shanghai off to a rival like BYD rather than protect it at all costs. He closes by noting the irony of Musk publicly calling the whole story fake news without ever simply stating that no merger is planned in the short, medium, or long term, a silence Eli reads as more suggestive than any denial.