The Orbital Layer: Why Space Became Infrastructure

The space economy hit $630 billion in 2025. Morgan Stanley, Bank of America, and the Space Foundation all converge on $1 trillion by 2030 and $1.8 trillion by 2035. Commercial activity now drives 78% of total revenue. Launch costs have fallen 95% in a decade.

And yet most investors still file space under "deep tech" -- a category that signals long timelines, uncertain returns, and science projects masquerading as businesses.

That framing is outdated. Space isn't an exploration thesis anymore. It's an infrastructure thesis. The real opportunity isn't rockets. It's the connectivity, compute, and data services that ride on top of satellite networks. Space is becoming the next cloud layer -- except it covers the entire planet, including the 90% of Earth's surface where ground infrastructure doesn't exist.

Direct-to-Cell Changes the Math

SpaceX launched Starlink direct-to-cell service commercially in July 2025 through a T-Mobile partnership. Starlink now has 10 million subscribers globally. At MWC Barcelona in March 2026, SpaceX rebranded the service to Starlink Mobile and announced the V2 satellite constellation -- larger birds with significantly more bandwidth, enabling video streaming and remote work from phones with no special hardware.

This is the moment satellite communications stopped being a niche product. When a standard smartphone connects directly to a satellite network, the entire telecom map gets redrawn. Rural areas, maritime routes, aviation corridors, disaster zones -- all become first-class coverage areas without a single new cell tower.

AST SpaceMobile is taking the same bet from the opposite direction: building its own constellation specifically for direct-to-cell, with $1.2 billion in revenue commitments from carrier partners. The competitive dynamics here are real, but the TAM validation is the point. Multiple billion-dollar companies are converging on the same thesis: the phone in your pocket will talk to satellites, and it will happen in the next 24 months at scale.

The question for carriers is existential. Does direct-to-cell destroy the telco business model or extend it? If Starlink Mobile delivers broadband from orbit, what is the value of a ground-based network in rural and suburban markets? And if carriers partner with satellite operators rather than compete, who captures the margin?

The Three Layers of Space Infrastructure

The investment landscape in space breaks into three layers with very different risk profiles and return characteristics:

The transport layer is largely won. SpaceX launches more mass to orbit than every other provider combined, and Falcon 9 reusability gives it unit economics no competitor can match on a relevant timeline.

The connectivity layer is where the next wave of massive value creation happens. Starlink is the incumbent, but this is a market that can support multiple large-scale operators -- the total addressable market is every person, vehicle, vessel, and device on Earth.

The compute and data layer is where things get genuinely strange -- and where the venture-scale opportunities live.

Compute in Orbit

Starcloud, a YC graduate, raised a $170 million Series A in early 2026 at a $1.1 billion valuation. The pitch: GPU-powered satellites that function as orbital data centers.

This sounds like a solution looking for a problem until you consider three things. First, terrestrial data center capacity is hitting real constraints -- power, permitting, water, NIMBYism. Hyperscalers are already exploring nuclear power just to keep building. Second, orbital data centers have zero land costs, zero water costs, and access to continuous solar energy. Third, for workloads that don't require ultra-low latency (training runs, batch processing, offline inference), orbit works.

Aalyria, spun out of a Google project, reached a $1.3 billion valuation building the network orchestration layer -- software that manages laser mesh networking between satellites, aircraft, drones, and ground stations. If orbital compute scales, someone needs to be the router. Aalyria is building that.

CesiumAstro closed a $470 million Series C in February 2026 -- $270 million in equity plus $200 million in debt -- backed by Woven Capital, Airbus Ventures, and the Development Bank of Japan. CesiumAstro builds software-defined communications systems for satellites. Think of it as the Qualcomm of space: the communications chipset and software layer that every satellite constellation needs.

Xona Space raised $92 million to deploy a LEO-based precision positioning constellation. GPS was built in the 1970s and operates from medium Earth orbit with meter-level accuracy. Xona delivers centimeter-level accuracy from low Earth orbit. The use cases are autonomous vehicles, precision agriculture, drone operations, and defense -- every domain where "close enough" positioning isn't good enough.

The Middleware Thesis

Most space investment discourse focuses on the infrastructure itself -- the satellites, the rockets, the ground stations. But the compounding value is increasingly in the middleware: the software and services that sit between raw satellite infrastructure and the applications that use it.

This is the same pattern that played out in cloud computing. AWS built the infrastructure. But Cloudflare, Datadog, Snowflake, and Stripe built the services on top that captured enormous value. The space equivalent is emerging: companies that take raw satellite connectivity, compute, or positioning data and turn it into application-ready services for autonomous driving, maritime logistics, precision agriculture, insurance underwriting, defense operations, and sovereign communications.

The companies building the Cloudflares and Stripes of space are where the venture-scale returns will concentrate. They're capital-light relative to constellation operators, they benefit from the infrastructure buildout regardless of which constellation wins, and they compound as the application layer expands.

The Investment Frame

The space economy is no longer speculative. It's a $630 billion market growing at 8% annually, with clear line-of-sight to $1 trillion. Commercial activity dominates. The infrastructure is being built by well-capitalized operators.

The transport layer is won. The connectivity layer is being contested. The compute and data layer is nascent. And the middleware layer -- the services that translate raw space infrastructure into application-ready capabilities -- is where the asymmetric opportunities live.

The question worth debating: if space is becoming a utility layer like cloud or fiber, who captures more value long-term -- the constellation operators building the pipes, or the middleware companies building the services that ride on top?