SpaceX is on the cusp of a record-setting initial public offering. Before that happens, management will try to convince investors to buy into founder Elon Musk's vision of the future.

It's best to hear directly from Musk. He spoke with potential JPMorgan investors at a Thursday event hosted by Jamie Dimon, who introduced Musk as the "Edison of our time."

Projections

The talk was long on vision, and short on details. That isn't a surprise. People familiar with Musk know he focuses on big goals that help motivate employees -- and investors.

He left the projections up to the bankers. Morgan Stanley says sales could reach $3.4 trillion in 2040, according to the Wall Street Journal, which cited an analysis shared with investors. SpaceX didn't respond to a request for comment about the figures.

Morgan Stanley and Goldman Sachs both project 2028 revenue of about $160 billion. NYU Professor and valuation maven Aswath Damodaran projects 2036 revenue of $420 billion.

Sales in 2026 might top $40 billion. Numbers are all over the place, which is one reason Musk's ability to share a coherent narrative matters.

Why Now?

SpaceX has been cash-flow positive for more than a decade, said Musk, but "we are embarking on a significant growth phase where we're going to put in orbit probably a 100,000 satellites."

Updated versions of SpaceX Starlink satellites have more than 100 times the bandwidth of existing versions. (Starlink is SpaceX's space-based broadband product with more than 10 million subscribers in 164 countries.)

Getting more bandwidth is critical to the AI revolution Musk sees, because robots and robo-taxis need ample bandwidth to operate successfully.

"We're also doing the AI data centers in space, which is another massive capital endeavor," added Musk. "But I think it will be the primary means by which AI can be expanded. It's increasingly difficult to build power plants on the ground."

Data Centers In Space

AI underpins SpaceX's proposed $1.8 trillion valuation. That makes space-based data centers critical to the future. The technology doesn't exist yet, but Musk doesn't seem worried.

"We don't think this is a particularly difficult thing to do. In Fact, we think it's easier than our communications satellites," said Musk. "The AI data center will be much simpler by comparison."

Starship

SpaceX's huge, fully reusable Starship is also key to SpaceX's future. It can reduce the cost of reaching space by 90% compared with a SpaceX Falcon 9, which has already cut the cost of reaching orbit by some 95% compared with the Space Shuttle.

"Once you achieve full reusability, [reaching space] is just the cost of propellant," said Musk. That's a little like commercial air travel. They don't throw the plane away after each flight.

Lower costs created the modern space economy, enabling communications, Earth imaging, and defense applications. There will be others.

Starship is also required to launch the third version of Starlink satellites, the ones needed to boost bandwidth.

Bottlenecks

Semiconductors are one of the bottlenecks to growth he sees: "What we see as a limiting factor is being able to make chips, both logic, memory, and [semiconductor] packaging."

It's why SpaceX is vertically integrating into chip production with Tesla at the "Terafab" in Texas.

Talking about chips, Musk slipped and said "Tesla AI satellite," which drew chuckles from the crowd. A few Wall Street analysts and investors see Tesla and SpaceX merging in the future.

Multiplanetary

Musk's long-term goal is to make human life multiplanetary. A moon base is an interim measure that will enable testing and provide a platform for sending supplies to Mars. It's easier to get things off the moon; it has less gravity than Earth.

"It would be pretty cool if you could vacation on the moon," Musk said.

Space hotels are a business investors haven't modeled yet.