EP74 — Unicorn Files — Open Cosmos: can Europe build a sovereign space network without copying Starlink?

Open Cosmos began in 2015 by simplifying the design, construction, launch and operation of customer satellites through OpenOrbit. Its strategy now links OpenConstellation shared Earth-observation capacity, DataCosmos processing and tasking, and ConnectedCosmos, a proposed specialised connectivity layer. The episode separates delivered customer missions and the January 2026 launch of two satellites from the targets of over 96 satellites by end-2028, a 192-satellite core and global coverage in 2030. We compare its focus on government and enterprise infrastructure with Starlink, Eutelsat OneWeb, and the EU IRIS² programme. The 5 November 2026 Liechtenstein notice confirms an ITU extension for filing 3ECOM-1 but not 3ECOM-3, highlighting deployment and regulatory risk. The €300M September financing is reported without a disclosed equity/debt split or official valuation; a roughly $2B value has been reported by the press, citing people with knowledge of the deal.

• Open Cosmos is moving from turnkey customer missions toward shared infrastructure, data and connectivity; mission delivery is demonstrated while the complete network is not.
• The €300M announced on 14 September 2026 is funding, with no public proof that it is all equity and no officially disclosed round valuation.
• The roughly $2B valuation comes from anonymous sources cited by the press and has not been confirmed by the company: Open Cosmos is a reported unicorn, not one with an official valuation.
• Two ConnectedCosmos satellites were launched in January 2026, while a core of 192, more than 96 by end-2028 and global coverage by 2030 are targets.
• The stated capacity to manufacture one satellite per day across four factories is not the same as actual output or proven economies of scale.
• On 5 November 2026 Liechtenstein announced that ITU extended the 3ECOM-1 M2 milestone until 30 June 2027 but declined an extension for 3ECOM-3.
• A European alternative to Starlink may not need comparable physical scale; the test is reliable, used and paid-for critical capacity.