SpaceX Chip Supplier: The Royalty Math in Jovine’s Pitch

The phrase “SpaceX chip supplier” is doing a lot of work in the current newsletter cycle, thanks to Dylan Jovine’s “Elon’s 100X AI Stock” presentation from Behind the Markets. The pitch’s core move is to argue that Musk’s orbital AI buildout has one dependency he cannot build himself, and that the company filling it collects a royalty on every unit of the fleet. Here is the supplier logic, the royalty math, and the honest breakdown of the opportunity, the risks, and who it suits.

Why Elon Licenses Rather Than Builds

The pitch sets up the dependency with a clean inventory of what Musk owns: “He has the rockets. He has the launch sites. He has the AI… There’s just one thing he can’t build himself. The chip that runs them.” The investment structure follows immediately: “Elon doesn’t own that chip. He licenses it. That means, every time a Tesla rolls off the line… Every time SpaceX launches an AI1 satellite… This company gets paid.”

Why not just build his own? The pitch’s answer is the switching cost: “Because leaving this company means rewriting every line of software in every Tesla and every satellite he owns… And it means giving up the only chip that can run AI in orbit without cooking itself.” That second clause matters more than the first. In orbit, a vacuum offers no convective cooling; waste heat can only leave by radiation, across radiator panels the pitch describes as “enormous” relative to the compute served. Power efficiency therefore gates how much AI a satellite can fly at all, which makes the most efficient architecture not a vendor choice but a physical requirement. That is the strongest version of the supplier argument, and it is sound engineering logic even where the marketing overstates.

The Royalty Math, From 10,000 to One Million

The scale story comes from a filed request the pitch cites: “SpaceX asks the FCC for permission to launch one million satellites,” versus “about 10,000 satellites in orbit” today. Hence the campaign’s name: “My research indicates that Elon’s bill to this firm could soon get a LOT bigger. 100X bigger.” The pitch also dates the relationship: “Elon has been quietly paying this obscure firm since 2012,” with the designs running Teslas, Starlink dishes, and AI1 satellites alike, and data center royalties that “more than doubled last year… And they more than doubled the year before that.”

Parsing the math honestly: a royalty-per-unit model does scale linearly with fleet count, so the inference is structurally coherent. But it assumes the fleet gets built at anything like the filed scale, and a filed request is a ceiling, not a schedule. The one outcome the pitch frames in dollars is modest by genre standards: “Even a 400% move from here would turn $5,000 into $25,000.”

The Fingerprint, Flagged

The company is unnamed; the name and ticker sit in Report #1, “Elon’s 100X AI Stock: The Chip Elon Can’t Fly Without,” included with the subscription. The pitch’s fingerprint: a chip-IP licensor collecting per-unit royalties across nearly every device category; a Tesla relationship since 2012; a history that includes “NVIDIA tried to buy it outright” before objections from Amazon and Samsung and a “national security concerns” kill from regulators; one giant shareholder still controlling “the vast majority of the stock” with “only a sliver” trading publicly; and institutional holders cited at $675 million, $349 million, and $343 million. We will say it plainly: the identification is unconfirmed. The clues point toward a chip-architecture licensor, and we treat that as a fingerprint, not a name. Any article asserting a specific company is going beyond what the pitch itself provides.

The Opportunity, the Risks, Who It Suits

The opportunity: a royalty business tied to the fastest-growing compute buildout in history, with lock-in reinforced by software switching costs and orbital thermal physics, plus a December 8th event that the pitch argues will redirect freed insider capital toward suppliers. The verifiable substrate holds: $725 billion of hyperscaler AI capex in 2026, a projected 49-gigawatt power shortfall by 2028, and a real orbital race (Starcloud trained the first AI model in space; Google’s Project Suncatcher is underway). See our orbital data center theme piece for the full map.

The risks: every company-specific claim is unverifiable before subscribing; the thin public float cuts both ways; the one-million-satellite scale is a request, not a deployment; and the lockup-expiry catalyst is genuinely debatable, since expirations are conventionally read as supply pressure. Jovine hedges with “I believe” and “in my opinion,” which we credit. We examined the December 8th event in our SpaceX lockup expiration analysis, and Jovine’s supplier method in our Superdeal stock pick coverage.

Who it suits: investors comfortable with a researched-but-unnamed supplier thesis, bought through a cheap, well-guaranteed wrapper ($49 per year, six-month full refund, keep the reports, A+ BBB rating). Not for anyone who needs the name before the thesis. If you want the argument in full, the presentation is here.

NewsletterVetter is an independent publication. We receive compensation from some of the services we review through affiliate links. Nothing on this site is investment advice. Always do your own research.