NASA is giving Relativity Space a 2028 Mars job before the company has reached orbit. That sounds reckless, but it also shows exactly how badly NASA wants more than one serious commercial path beyond Earth.
NASA's June 17 announcement is simple on paper and risky in practice: Relativity Space will supply the spacecraft, rocket, and cruise operations for Aeolus, a Mars orbiter scheduled to launch in 2028. NASA will supply the science payload, a four-instrument suite meant to give researchers the first integrated daily global view of Martian winds, temperatures, dust, and clouds.
Here is the part that should stop you. Relativity has never put a payload in orbit. Its only launch, Terran 1, lifted off from Cape Canaveral on March 23, 2023, survived Max Q and stage separation, then failed after the second-stage engine cut out. The rocket fell into the Atlantic. Relativity retired Terran 1 soon after and moved its attention to Terran R, a much larger partially reusable medium-to-heavy-lift rocket that still has not flown.
That is not a small caveat. According to NASA, Aeolus is meant to carry a Doppler Wind and Temperature Sounder, a Thermal Limb Sounder, a Surface Radiometric Sensor Package, and a Wide-Field Context Camera. The payload is being designed, built, and integrated at NASA's Ames Research Center in California's Silicon Valley, while Relativity manages spacecraft development and mission operations. NASA also says it will support science instrument operations for at least one Martian year.
So no, this is not just a branding win for Eric Schmidt's rocket company. It is a real planetary science mission with a real clock attached.
Schmidt became Relativity's CEO in March 2025 after taking a controlling stake in the Long Beach company, replacing co-founder Tim Ellis in the top job. That move looked odd from the outside. AI infrastructure was swallowing capital, Blue Origin was still trying to prove New Glenn at scale, and SpaceX was making the launch market look brutally hard for everyone else. But Schmidt appears to be making a colder bet: if compute is one layer of the next infrastructure race, access to orbit is another.
Relativity gives him a company with ambition, customers, and one very public miss. TechCrunch reported that NASA has not disclosed how much it is paying Relativity for Aeolus, and that Relativity did not respond to its questions. That dry fact matters. The company is not being carried by a giant public contract with every detail spelled out. It is taking on development risk in the style of NASA's commercial partnerships, where the agency brings the mission need and the private company absorbs part of the cost in exchange for a marketable capability later.
NASA has used that pattern before. SpaceX's cargo work for the International Space Station turned a risky startup relationship into a working logistics line. Firefly Aerospace has taken NASA payloads to the Moon under a similar commercial model, with all the messiness that comes when young companies try to do hard things on public deadlines. SpaceX failed its first three Falcon 1 orbital attempts before reaching orbit in 2008. If you only buy proven space hardware, you get fewer new providers.
Frankly, that is the strongest argument for the Aeolus decision. NASA does not need another slide deck about commercial competition. It needs rockets that fly, spacecraft that work, and companies that can take responsibility for missions beyond low Earth orbit. Relativity cannot become that company unless someone gives it work that requires it to grow up.
The SpaceX shadow sits over the whole story. Elon Musk's company has not sent its own private mission to Mars, as TechCrunch pointed out, but it dominates the American launch conversation in a way no government agency should find comfortable forever. NASA can admire SpaceX's execution and still want credible alternatives. Those two ideas are not in conflict. If anything, a second serious provider gives NASA more leverage on price, schedule, and political risk.
Mars, though, is a cruel place to test a hopeful theory. Launch windows open roughly every 26 months when Earth and Mars line up favorably. Aeolus is targeting 2028. Miss that window and the mission waits until 2030. That leaves Terran R very little room for a messy debut, a long redesign, or a run of flights that proves only that the rocket needs more time.
That is the wager. Schmidt needs Terran R to become more than a funded promise. NASA needs more commercial capacity without pretending unflown rockets are already mature. Aeolus sits right between those needs, carrying Mars weather instruments on one side and an industrial test on the other.
You should not read this as NASA losing discipline. You should read it as NASA deciding that dependency has a cost too. Relativity may fail to meet the 2028 window, and if Terran R stumbles, the agency will have to answer for trusting an unproven rocket with a Mars mission. But if it works, NASA gets more science, Schmidt gets proof that Relativity can fly serious missions, and the launch market gets one less excuse to behave like SpaceX is the only answer.
Also read: Unilever and Accenture bet on digital twins to prove industrial AI pays off • Bland raised $50 million after 180 investors said phone calls were a dying medium • Singapore is becoming the AI world's Switzerland, but China just showed the escape hatch has limits
