Analysis
Venture funding into space startups has hit a record $20.3 billion so far in 2026, with orbital compute emerging as a new investment category alongside the launch and satellite-manufacturing bets that have historically dominated the sector, Tech Times reported.
The record comes in a week when Pulse also covered Isar Aerospace's second orbital launch attempt from Norway -- the kind of launch-vehicle milestone that has anchored space investing for a decade, even as capital increasingly diversifies into adjacent categories.
Smaller Deals, Broader Base
Unlike the multibillion-dollar AI infrastructure rounds dominating this week's headlines, space capital is still deploying mostly through smaller, more numerous checks:
- Kepler Aerospace (Bengaluru, India) -- $8 million seed led by Blue Ashva Capital, funding the launch of its first six satellites and a mission-control facility expansion.
- Insera Solution (South Korea) -- roughly $9.7 million (₩13.5 billion) Series A from SV Investment, Shinhan Venture Investment and Intervest, building precision laser-pointing systems for inter-satellite communications.
The Orbital Data Center Thesis
The newest wrinkle is orbital compute: startups pitching data centers in orbit, powered by continuous solar exposure and cooled by the vacuum of space, as a longer-term answer to the same power and land constraints driving terrestrial AI infrastructure deals like the ones Pulse covered in this week's lead. The category is unproven at any commercial scale -- no orbital data center has moved meaningful compute workloads yet -- but its emergence as a distinct funding line item shows investors treating space-based infrastructure as a genuine extension of the terrestrial AI buildout rather than a separate, unrelated thesis.
The practical read for space-focused GPs: capital is broadening beyond the launch vehicles and satellite constellations that defined the last funding cycle, into ground -- and now orbital -- infrastructure layered on top of them. That is the same maturation pattern the terrestrial cloud market went through roughly fifteen years ago, compressed into a much shorter timeline because AI compute demand is pulling multiple adjacent categories forward at once.
