The Reckoning Sequence
Musk became the world’s first trillionaire. Anthropic filed for a near-trillion-dollar IPO. And somewhere in the middle of all that, SpaceX paid $60 billion for a coding tool that’s losing market share. This is what the infrastructure war looks like when it goes public.
Part One: The Throne and the Receipt
On June 12, 2026, Elon Musk became the world’s first trillionaire. SpaceX debuted at $150 per share, raised $75 billion in the largest IPO in history, and briefly crossed a $2.5 trillion market cap. The crowd went wild.
The numbers behind the crown were less flattering. SpaceX’s IPO priced at roughly 94 times revenue — compared to Meta at 22x and Amazon at 18x at their own debuts. To justify that valuation, analysts noted, SpaceX’s 2035 earnings would need to grow 75-fold from 2025 levels. xAI posted a $6.36 billion operating loss in 2025 on $3.2 billion in revenue. X generated about $1.8 billion in ad revenue last year — less than half of what Twitter earned in 2021 before Musk bought it for $44 billion.
The world’s first trillionaire built his throne on a valuation that requires magical thinking — then immediately spent $60 billion trying to buy his way into a race someone else already finished.
Five days after the IPO, SpaceX announced it would acquire Cursor for $60 billion in stock. The deal was framed as a power move into AI development tools.
Here’s what the headlines didn’t lead with: Cursor’s market share in AI coding tools dropped from 41% in June 2025 to approximately 26% by May 2026. Anthropic now controls roughly half that category. SpaceX paid a trillionaire’s ransom for a product in structural decline, in a category its primary rival already dominates.
That’s not a power move. That’s a correction disguised as an acquisition.
Part Two: The Infrastructure War (A Recap)
Six weeks ago, I wrote a two-part research brief for this newsletter. The thesis: while the public watched the chatbot horse race, a quieter competition was underway for the infrastructure layer of AI — the runtimes, SDKs, developer tooling, and enterprise pipelines that would determine who controlled how AI got built and deployed.
OpenAI’s strategy: buy access. Wide surface area, high velocity, platform logic.
Anthropic’s strategy: buy control. Fewer acquisitions, each one load-bearing. The plumbing, not the faucet.
Meta’s strategy: become the training data. Not a new strategy — just one they’ve stopped pretending isn’t what it is.
The Cursor deal is the receipt for that thesis. xAI had resources and ambition. What it apparently lacked was the product. So SpaceX wrote the check.
The problem is that Anthropic didn’t just win the developer tools category by having a better product. It won by building the category around itself. Claude Code. Enterprise integrations. The SDK ecosystem. By the time SpaceX wrote the $60 billion check, the infrastructure layer was already poured.
You can buy a tool. You can’t buy the ecosystem that already formed around a competitor’s tools.
Part Three: The IPO Triple Play and What It Actually Means
Meanwhile, three of the most consequential public offerings in history are converging on the same six-month window.
SpaceX debuted in June. Anthropic filed its confidential S-1 on June 1, targeting a valuation of approximately $965 billion, with a potential listing as early as October 2026. Revenue run-rate hit $47 billion in May — up from $9 billion at end of 2025. Enterprise market share surpassed OpenAI’s for the first time in April. OpenAI filed its own confidential S-1 days later, targeting a September debut at $730-850 billion.
Perplexity CEO Aravind Srinivas put it plainly: “I certainly think there will be ripple effects if they don’t go well — there is no sugar coating on that. The SpaceX IPO will definitely be a leading indicator for how Anthropic or OpenAI will go out.”
Perplexity is watching from 2028. Deliberately. Their chief business officer framed it as strategy: “By consistently holding 2028 as our earliest date for an IPO, Perplexity has been able to build a healthy, high-growth business.” That’s not patience. That’s a calculated read on what happens when you go to market before the category fully clarifies.
“The SpaceX IPO will definitely be a leading indicator for how Anthropic or OpenAI will go out. There is no sugar coating on that.” — Aravind Srinivas, CEO, Perplexity
The stakes are structural. If SpaceX’s valuation holds and the Cursor acquisition reads as strategic, it sets a floor for AI infrastructure multiples. If the market looks at 94x revenue and blinks, the whole sequence compresses.
What makes this moment different from previous tech IPO cycles: these companies aren’t pre-revenue. Anthropic’s annualized revenue grew more than fivefold in five months. The question isn’t whether the businesses are real. It’s whether the multiples price in a world where AI compounds indefinitely — or one where it turns out to be a slightly more sophisticated version of the last wave of software.
Part Four: The User Test
Here’s a data point no analyst will put in their model.
I subscribed to Cursor about a week ago. I use Claude Code for deep work — complex reasoning, architecture decisions, anything that requires sustained context. I use Cursor for fast tasks. Speed. Iteration. The quick thing that doesn’t need the full weight of Claude behind it.
That split is deliberate. And it’s the split that makes Cursor valuable to me. Not because Cursor is better than Claude Code. Because it routes intelligently between models depending on what the task needs. The optionality is the product.
SpaceX paid $60 billion for a customer base that will cancel the day the product gets worse. That’s not an acquisition. That’s a lease with a very expensive deposit.
The moment Cursor becomes a Grok delivery mechanism — locked to xAI’s models, optimized for SpaceX’s infrastructure ambitions rather than developer outcomes — that value proposition collapses. Not because Grok is terrible. Because the optionality disappears.
And Cursor’s audience will notice immediately. These are developers. Deeply opinionated about tooling. They already voted once — that’s how market share fell from 41% to 26% in under a year. The audience is pre-churned and paying close attention.
Musk is betting they’ll stay because switching costs are real and Grok will get good enough fast enough. He might be right. But the window is narrow, and developer loyalty has a very short half-life when the tool stops delivering.
The infrastructure play only works if the product stays good enough to keep the users who justify the infrastructure. Right now, that’s not guaranteed.
What the Chaos Actually Translates To
Here’s the pattern I keep seeing, and it’s the same one I wrote about six weeks ago:
The companies that move on infrastructure early — quietly, without press releases — end up owning the category. Anthropic didn’t announce it was going to dominate enterprise AI coding. It built Claude Code. It built the SDK ecosystem. It shipped. And then, in April 2026, its enterprise market share crossed OpenAI’s for the first time.
SpaceX, sitting on a $2 trillion market cap and a trillionaire CEO, had to write a $60 billion check to try to catch up. That’s not winning. That’s the cost of watching while someone else built the foundation.
The IPO race that’s coming will answer a specific question: does the public market understand the difference between a company that won the infrastructure layer and a company that’s still fighting for position in it? Srinivas thinks the SpaceX IPO is the leading indicator. I think the Cursor acquisition is.
A company that just became the most valuable in its sector by a historic margin immediately spent $60 billion trying to buy into a category it should have owned. That’s the tell.
Watch where Anthropic prices in October. That’s when we find out if the market read the same signal.
If this landed — forward it to someone who needs to read it.