The post-ISS era is going to be weirder than you think

Two weeks ago I wrote that the launch cost curve had bent. This week I want to look at what happens because of it. The most important single event in commercial spaceflight for the rest of this decade is going to be the end of the International Space Station. It is going to be messy, it is going to be late, and almost every prediction about it that you've read is going to be wrong in a specific way. I want to walk through what's actually known, what's actually unknown, and why I think the period from 2027 to 2032 is going to be the strangest stretch of orbital operations since the Shuttle-Mir program.

The deadline that's not quite a deadline

The headline version of this story is simple. The ISS is going to be deorbited at the end of 2030. NASA says so. The other partner agencies say so. The plan, per the NASA ISS Transition Plan and confirmed in the most recent GAO review (GAO-26-107805, "Low-Earth Orbit: NASA Faces Impending Decisions"), is to hand off human presence in low Earth orbit to commercial operators and then bring the station down in a controlled re-entry over a remote ocean stretch (Source: NASA Transition Plan; GAO-26-107805).

The story is more complicated than that. Russia, one of the five original ISS partners, has committed only through 2028, not 2030 (Source: USA Today, citing a 2023 NASA blogpost). The Russian segment — currently the only place on the station that can execute a controlled deorbit burn using the station's own propulsion — is, in the strict sense, the segment that determines the deorbit timeline. If Russia pulls its cosmonauts and modules in 2028, the deorbit has to happen earlier than 2030, or NASA has to find another way to bring the station down. (The contingency is a US-built deorbit vehicle; SpaceX has the contract for it. It is in development. It is not in orbit.)

So the deorbit deadline is closer to "end of 2030 if everything works, end of 2028 if it doesn't, and some time in 2031 if NASA pushes an extension through the structural concerns flagged by NASA's Inspector General in late 2024." That uncertainty window is enough to break the clean-handover narrative.

The four contenders, and what they actually are

There are four credible US commercial space station efforts. They are not all the same kind of thing, and confusing them is the single biggest source of bad coverage on this topic.

Axiom Space. The only one with a current flight contract. Axiom's plan is to launch modules that attach to the ISS first, then detach as a free-flying station. In June 2024, the company revised its assembly sequence so that the Power, Thermal, and Payload module launches first, allowing the complex to separate as a free-flyer "as early as 2028" (Source: Axiom Space press release). This is the only path that uses the ISS as an anchor — which is both its biggest advantage and its biggest risk, because it inherits the ISS retirement timeline.

Vast Space. Founded in 2021 by Jed McCaleb, Vast is going to be the first commercial company to put a station in orbit if it holds its current schedule. Haven-1 was originally targeted for 2026, slipped to May 2026, then slipped again on January 20, 2026 to "no earlier than Q1 2027" (Source: Vast announcement, reported by Payload, Aviation Week, Universe Magazine). The station is roughly 14,000 kilograms, designed to launch on a single Falcon 9, with a three-year service life supporting up to four short-duration crewed missions via Crew Dragon (Source: Space.com, Vast). Vast has also announced Haven-2, a NASA-CLD-bid evolution of Haven-1 that it claims could be operational by 2028 if selected in 2026 (Source: Vast press release at IAC 2024; Space.com).

Voyager / Starlab. A joint venture of Voyager Technologies, Airbus, Mitsubishi, and MDA Space. Single-launch architecture on SpaceX Starship, target 2029 (Source: Airbus product page, Voyager, Wikipedia). Starlab's schedule is directly tied to Starship's cadence, which is itself a moving target. The Voyager Space Act Agreement from NASA's Phase 1 program was $217.5M; $147M of that has been paid out as of the most recent reporting (Source: Kavout, citing NASA filings).

Orbital Reef. A Blue Origin / Sierra Space joint venture, with the Sierra-built LIFE inflatable module and the Dream Chaser cargo vehicle as central pieces. In June 2025 the project completed its System Definition Review with NASA (Source: Sierra Space press release). Sierra's published target for operational capability is 2027 — that is a Sierra projection, not a NASA contract milestone, and I want to flag it as such.

There is also a long tail of smaller efforts — Northrop Grumman (which exited Phase 2 in late 2024 but is still in the CDISS pool), Think Orbital, Special Aerospace Service, SpaceX's standalone habitat work (quietly rumored). None of these are currently credible as a 2030 ISS replacement. They are interesting for other reasons.

What the schedule actually looks like

Here is where I want to push back on the prevailing narrative. The dominant story is "the ISS ends in 2030, and commercial stations take over." The actual story is much weirder.

The first commercial station in orbit will almost certainly be Vast Haven-1, if Vast holds its Q1 2027 launch date. That is not an "ISS replacement." It is a single-module, single-launch demonstrator with a three-year service life, designed to support short crewed missions and to prove that a private company can independently operate a habitable spacecraft.

The first NASA-certified "commercial destination" — the first station that can host NASA astronauts for the long-duration research the ISS currently supports — is not on any published firm schedule. NASA restructured its CLD Phase 2 acquisition strategy in August 2025 (Source: NASA directive dated August 4, 2025, "Directive on Revised Commercial LEO (LEO) Destinations (CLD) Phase 2 Acquisition Strategy"). The current best public information is that NASA intends to select one or more destinations in 2026, with operational capability required in the 2028-2030 window — but those are program targets, not contract milestones.

So the actual timeline, as best I can assemble it: Vast Haven-1 launches early 2027 (best case) or slips further. Axiom's free-flyer separates from the ISS as early as 2028. Starlab launches on Starship, no earlier than 2029. Orbital Reef becomes operational "by 2027" per Sierra Space's projection, but I would not bet on it. One or two NASA-certified destinations emerge from the CLD Phase 2 competition. The ISS deorbits in late 2030 (or 2031, or sooner).

That gap between "first commercial station in orbit" and "ISS gone" is the part of the story I want to spend time on. Because that gap is where the weirdness lives.

What the weirdness actually looks like

The clean-handover narrative assumes that commercial stations will be ready to absorb ISS-class research demand the moment the ISS is gone. That is not what the public schedule supports. What is more likely is a transitional period in which:

• Multiple stations operate simultaneously, each at a different scale and capability. A single small commercial module in 2027 is not the same category of asset as the ISS. But it is the first operational US commercial station, and it is going to get treated as a "first" by the press even though it isn't an ISS replacement.
• The ISS runs to 2030 (or 2031) under a controlled, declared extension, because NASA and the partners are going to be reluctant to deorbit the only proven platform for long-duration microgravity research before an equivalent capability is demonstrably in place. The GAO has flagged this as a risk; NASA's OIG has flagged it; Congress is aware of it.
• The first "ISS replacement" station is going to be smaller than the ISS. All four contenders are. Starlab is the only one designed to launch as a single integrated complex on a single launch. Axiom's free-flyer is a modular build-up. Vast's Haven-1 is a single module. Orbital Reef grows incrementally. None of them will match the ISS pressurized volume at first launch.
• The deorbit itself is going to be a story. The ISS mass is around 420,000 kilograms. Bringing it down safely requires a deorbit burn too large for the station's own propulsion — hence the SpaceX USDV (US Deorbit Vehicle) contract. That vehicle is in development. It is not guaranteed to be ready in time.

What this means in plain English: there is a non-trivial probability of a 12-to-24-month period in which the only continuously-crewed US orbital platform is one or more of these early commercial stations, and the first NASA-certified "destination" station has not yet flown. That period will be characterized by overlapping capability gaps, ad-hoc mission extensions, and pressure on the deorbit decision.

What to watch for

If you want to actually follow this story as it develops — and you should, because it is the most consequential single thing happening in human spaceflight — these are the specific events to track.

Vast Haven-1 launch, Q1 2027 or later. A successful launch puts the first commercial station in orbit and reframes the entire CLD competition. A further slip pushes the whole timeline right.

NASA CLD Phase 2 awards, 2026. The formal selection of one or more commercial destinations for NASA certification. This is the decision that turns the program from "funded demonstrators" into "the actual ISS replacement."

The USDV deorbit vehicle timeline. The SpaceX-built deorbit vehicle is the single point of failure for the 2030 deorbit plan. Any meaningful slip forces a re-think of the whole transition.

The Russian segment decision, 2027 or 2028. Whether Roscosmos continues operations past 2028 determines whether the ISS can physically reach its planned 2030 deorbit under Russian-segment propulsion.

The first CLD-certified station operational capability. Whenever this happens, it will be the moment the commercial LEO era is real, not aspirational.

That last item is the one I am personally watching most closely. Everything else is schedule work. The operational capability date is when the rhetoric about commercial space stations stops being rhetoric.

If you want to read the closest thing to a serious book on the orbital economy as it stands today — and this is the period where that economy actually starts to become measurable — the Wiley title linked below is the one I'd hand to a smart friend who wanted to understand the commercial case for what's being built. It was published before the current CLD Phase 2 reshuffle, so some of its projections are dated, but the framework holds up.

And on a completely different note — the part of this story that the headlines miss is that the ISS is going to be visible from your backyard for the next four years. It is the brightest moving object in the night sky, and there is a free NASA service (Spot The Station) that emails you when it is about to pass over your location. If you have never watched it go by, the next four years are your last chance with this particular spacecraft. A pair of 15x70 astronomy binoculars is the most cost-effective way to actually see structure on it — the truss, the solar arrays, the modules — as it passes. It's not the same as a photo through a tracking scope, but it is a real, immediate way to participate in the end of an era.

That's it for this week's essay. Next Saturday: a deep dive on the orbital debris problem and why it's the thing that will quietly determine which of the commercial stations actually works.

— Atlas Renner, Editor-in-Chief, SpaceOrbitals