K2 Space's $500M Series D: A Contrarian Bet on Bigger Satellites
K2 Space, a Los Angeles-area satellite manufacturer, has raised a $500 million Series D at a $6.8 billion valuation, betting that the future of space belongs not to ever-smaller satellites but to big, high-power ones. The round, announced Thursday, was led by Kleiner Perkins and ICONIQ, with participation from Alphabet's CapitalG, Lightspeed, and ARK Invest. The new valuation more than doubles the roughly $3 billion figure the company carried just months ago.
The company's pitch is deliberately contrarian. A wave of startups has spent the past decade miniaturizing satellites to cut launch costs. K2 argues that in orbit, electrical power is the real constraint: a satellite with 20 kilowatts of power can host far more capable payloads than a typical small sat can. That logic underpins its first mega-class satellite, Gravitas, which reached orbit in March on a SpaceX Falcon 9 rocket.
Gravitas weighs about two tonnes, generates 20 kilowatts, and carried a dozen customer payloads, including equipment for the US military. K2 also plans to move from single satellites to production volumes: its 180,000-square-foot factory in Torrance, California is designed to build up to 100 satellites a year, with more than 85% of each satellite manufactured in-house.
The company has now raised more than $1 billion in total, against more than $1 billion in signed contracts. Its next milestone is the Trinity mission in early 2027, when multiple satellites are due to fly, followed by an even larger 100-kilowatt vehicle, called Giga, planned for late 2028 as a step toward carrying massive computing loads in orbit.
Inside K2's Strategy: Power, Scale, and the Military Pull
Why Going Big Makes Sense for K2
The satellite industry's default assumption is that smaller equals cheaper. K2's counterargument has real engineering logic: high-power satellites can do things small ones cannot, from heavy sensing payloads to, eventually, onboard computing. By building more than 85% of its own hardware, K2 also captures more margin per satellite and controls its production schedule — a vertical-integration model closer to SpaceX's approach than to that of many satellite bus makers.
Military Demand Is the Near-Term Engine
Much of K2's contracted backlog comes from defence customers. The company will supply 30 satellites for SES's meoSphere network, build spacecraft for the Anduril-led Golden Dome missile-defence programme, and act as bus provider for a US Space Force satellite communications effort — its first Pentagon programme of record. These are named, signed commitments with concrete delivery timelines. The US military's growing appetite for proliferated, resilient constellations is a structural tailwind for any manufacturer that can prove it can deliver satellites at scale and speed.
The Long Wager: AI Data Centres in Orbit
Beyond defence, the most speculative part of K2's story is the plan for a 100-kilowatt Giga satellite in late 2028, which CEO Karan Kunjur has described as a step toward putting massive amounts of compute in orbit. That would be a move toward true space-based AI data centres. The appeal is obvious: abundant solar power and no terrestrial land or cooling constraints. The obstacles are equally obvious — radiating heat in a vacuum, servicing hardware hundreds of kilometres up, and the sheer cost of launching and maintaining heavy computing infrastructure. This part of K2's vision is years away and unproven, so it should be treated as an option, not a near-term business.
The Manufacturing Challenge Is the Real Test
K2's ambition to build up to 100 large satellites a year would be a step-change in production for a class of vehicle that has historically been built in small batches. No one has yet pulled off mass production of two-tonne satellites. The company's progress through the 2027 Trinity mission and the ramp toward Giga will be the practical evidence of whether that scale is achievable. Until those milestones pass, the valuation premium assumes execution that has not yet been demonstrated.
What K2's Raise Signals for Space Investors and Contractors
For investors and industry observers:
- Watch K2's delivery of the 30 satellites for SES's meoSphere network and the Golden Dome spacecraft work — these contracts are the first real proof of K2's ability to scale beyond a single demonstrator.
- Track the Trinity mission in early 2027 as the next operational milestone; a successful multi-satellite flight would validate the factory's 100-per-year capacity claims.
- Monitor the 2028 Giga 100-kilowatt satellite as a long-dated catalyst, but treat space-based AI compute as speculative until a demonstration flight exists.
- For competitors and customers in the defence satcom market, K2's Pentagon programme-of-record status signals a third credible US bus supplier alongside established primes — a factor worth factoring into procurement timelines and pricing.
Risk & Opportunity Assessment
| Commercial Risk | High | K2 has over $1B in signed contracts but must scale from one launched satellite to producing up to 100 large satellites per year; any manufacturing delays could erode the $6.8B valuation and backlog confidence. |
| Competitive Risk | Medium | Most satellite startups focus on small sats, so K2 faces less direct competition in the large, high-power segment — but SpaceX, established primes, and potential defence-platform entrants could crowd the niche if military demand grows. |
| Regulatory Risk | Medium | Much of K2's backlog is tied to US military programmes (Golden Dome, Space Force satcom), making revenue concentrated in a single national security customer whose budgets and priorities can shift with political leadership. |
| Reputation Risk | Medium | The company has publicly promised a manufacturing capability no one has demonstrated, plus a speculative 2028 orbital-compute product; missed milestones could damage credibility with investors and Pentagon customers alike. |
| Technology Disruption | High | The 100-kilowatt Giga satellite and in-orbit AI computing push the limits of power generation, thermal management, and servicing; if those systems fail early, the long-term growth story weakens materially. |
| Commercial Opportunity | Transformational | With over $1B in contracts and a boom in defence-space funding, K2 is positioned to own the high-power segment and potentially pioneer orbital compute, a market that could redefine where AI workloads run. |
Comments 0