Weekly Recap: Weapons in Orbit, Money on the Table

Five point nine two billion dollars. That's the total value of NASA's Commercial Crew contract with SpaceX after this week's add-on — and it's just one piece of a week where money and hardware collided from low Earth orbit all the way to the geostationary belt. This is the weekly recap: the arc of the week in the business of space, story by story. Before we get into the money, a genuine first — the Pentagon publicly admitted, for the first time, that it has weapons parked in orbit. And underneath the headlines, there's a wave of institutional cash moving: a Rocket Lab raise worth nearly two billion dollars, a new Space Force surveillance contract, and a fight over where a brand-new Space Academy actually lands. Welcome back to Space Stakes, your daily brief on the business of space. It's Saturday, September 19, 2026. Let's get into the week that was.

Let's start with the number I opened on. NASA modified its Commercial Crew contract with SpaceX this week, adding three more missions — Crew-15, Crew-16, and Crew-17 — flying astronauts to the International Space Station. The modification itself is worth nine hundred forty-six million dollars, and it's structured as a firm fixed-price deal, meaning SpaceX, not the taxpayer, eats any cost overrun on those flights. That pushes the total value of NASA's CCtCap contract with SpaceX to five point nine two billion dollars. On X, NASA's own Space Operations account confirmed the mechanics directly, writing: "To continue regular crew transportation to the @Space_Station, NASA has awarded @SpaceX three additional missions through a contract modification. As part of the agency's Commercial Crew Transportation Capability (CCtCap) contract, this change brings the total missions for SpaceX to 17 and helps @NASA to maintain access to the space station with two unique commercial crew industry partners." That line about two unique partners matters, because the same week NASA also moved to keep Boeing's Starliner alive — more on that in a minute. So is nine hundred forty-six million dollars for three flights a good deal? Honestly, we can't fully check the math: NASA hasn't broken out the per-mission price, hasn't named crews, and hasn't set exact launch dates for these three missions. What we can say is the contract type protects the public purse — fixed-price means SpaceX bears the risk, not NASA. And that's the standard I think every big contract winner should be held to: if you're carrying this much of NASA's crew manifest, the public deserves cost discipline on the receipt, even when the receipt is short on detail.

Now, sticking with hardware — SpaceX spent the week building toward Starship's first attempted orbital flight, and the target date has already moved once. Flight 14 was originally pointed at next week; by midweek that had slipped to September 22; the newest word from NASASpaceFlight.com has it no earlier than September 28, pending regulatory approval. That's a real slip, not a rumor, and it's worth holding to the show's own standard — a date is a target until the vehicle actually leaves the pad. The one piece of hardware progress that did land: SpaceX completed chopstick-catch testing — using the launch tower's arms to grab the returning booster — with an unflown ship. That's a rehearsal, not a mission, but it's exactly the kind of ground-systems work that has to happen before an orbital attempt can happen at all. So does Flight 14 actually fly on September 28? I wouldn't bet the show on it — this date has already moved once this week alone.

Speaking of second providers — earlier this week Ars Technica reported NASA is moving to order more operational missions for Boeing's Starliner capsule, and to help certify a new rocket to keep it flying. That's a real turn for a program that had been effectively sidelined after years of technical trouble. The newcomer version: Starliner is Boeing's crew capsule, built as NASA's second, independent way to reach the space station alongside SpaceX's Dragon. What's new this week is NASA leaning in rather than quietly parking the program. But the open question from earlier coverage hasn't been answered — how much new money is NASA actually putting toward those thruster fixes and that rocket certification? Neither NASA nor Boeing has published that figure, and Starliner still hasn't flown an operational crew rotation mission. Pair that with this week's SpaceX award, and you get NASA's real posture: pay for redundancy, even when it's expensive and unproven, because a sole crew provider is a risk NASA doesn't want to carry.

For the first time, the US military has publicly confirmed it has weapons in orbit. Air Force Secretary Troy Meink said as much at the Air & Space Forces Association's Air, Space and Cyber Conference this week, confirming the country has orbiting space control weapons capable of defending the joint force. That's a genuine first — decades of official ambiguity, ended in a single public statement. What we don't have is any detail on what those systems actually are, how many exist, or what they can do — this was an acknowledgment, not a technical briefing. So here's the honest read: the disclosure itself is the news. The capability, apparently, already existed. What changed this week is that the government stopped pretending otherwise.

Rocket Lab had one of its busiest weeks of the year, and it cuts two directions at once. On the financing side, the company closed a one point nine four four billion dollar equity raise this week to fund its acquisition of Iridium — one of the largest financing moves in the sector this year. SpaceNews confirmed on X that Rocket Lab is on track to complete that Iridium acquisition. On the government side, the same company is protesting a loss: Rocket Lab filed a formal challenge after NASA picked Blue Origin for a seven hundred million dollar Mars orbiter contract instead. Those are two separate facts on two separate tracks — the equity raise closed, full stop, while the protest is still sitting with the GAO, unresolved. It's worth keeping straight, because a company can be simultaneously scaling aggressively through capital markets and fighting hard for government dollars it didn't win, and neither outcome tells you how the other one ends.

Here's one that flew under the radar this week: MDA Space unveiled a new Earth-observation product line called CHORUS, built around a constellation that hasn't actually launched yet. The portfolio comes in three tiers. CHORUS Command lets customers search MDA's existing Earth-observation archive directly. Command Pro goes further — install a regional ground station and get direct downlink, meaning imagery comes straight to the customer instead of routing through MDA's own systems. And CHORUS Intelligence is the finished product: things like vessel detection, sold as ready-made analysis rather than raw imagery. On X, MDA Space's own account called it "an end-to-end MDA CHORUS Earth observation product portfolio... this new suite lets customers customize, manage and control their geointelligence workflow." That's the company's own framing, worth taking at face value as far as it goes. The constellation behind all this — one satellite built by MDA, one by ICEYE, joining the existing RADARSAT-2 — is set to launch later in 2026, not yet in orbit. So MDA is selling the software and the product tiers ahead of the hardware actually flying. That's not unusual in this business, but it's worth naming plainly: what you just heard describes capability MDA intends to deliver, not capability it has delivered yet. The real test is whether that constellation launches on the stated late-2026 timeline, and whether the vessel-detection products actually work once there's real data behind them instead of a press deck.

Another one worth a closer look: the Defense Innovation Unit selected Northrop Grumman and True Anomaly — a startup paired with an incumbent prime — to build satellites called GHOST-R, meant for close-range reconnaissance in geostationary orbit, the ring roughly thirty-six thousand kilometers up where a lot of high-value military and communications satellites live. This feeds into the Space Force's broader RG-XX program, a multibillion-dollar effort aimed at fielding a fleet of GEO surveillance satellites by the end of the decade. Here's the taxpayer question worth asking: what does this actually replace, and does it save money? The stated goal is driving down lifecycle costs versus the current GSSAP satellites, which are expensive and built in low volume. But this is a prototype award, not an operational satellite — these aren't expected to launch until 2028, and a successful prototype still has to survive the jump to what the Pentagon calls a program of record before it becomes anything real. Pairing a newer company with an established prime is a pattern worth noticing across this year's defense contracts — one way the Pentagon is trying to keep competition alive instead of defaulting to the same handful of primes every time. Whether GHOST-R actually beats GSSAP on cost per satellite is a number nobody can check yet, because nobody's flown the prototype.

Sticking with space-domain awareness — the military's term for knowing what's actually up there and what it's doing — Colonel Barry Crocker, who heads Space Force's Mission Delta 2, said this week that improving support to space control operations is his unit's top priority, and that the service is moving to prototype new software that fuses commercial and military space-tracking data. Think of it as the layer meant to tie together everything else in this week's news: the newly disclosed orbital weapons, the GEO surveillance satellites Northrop and True Anomaly are building, all of it needs a system underneath that can actually make sense of what commercial radar and telescopes see alongside what the military's own sensors see. What we don't have yet is a contract, a dollar figure, or a delivery date — this is a stated priority and a prototyping effort, not a signed program. Which commercial data providers actually feed into this fused system, and under what contract vehicle, is exactly the kind of detail that turns a priority into a purchase, and it's not published yet. It's a genuinely boring-sounding story next to weapons in orbit or a two-billion-dollar acquisition, and that's exactly why it's worth covering — the plumbing decisions about whose data counts and how it gets stitched together end up shaping which companies get paid and which government systems become obsolete.

Last story of the week, and it's a pure political-geography fight. Back on August 28th, President Trump signed an executive order establishing a US Space Academy, and this week multiple states started publicly campaigning to be chosen as the host site. That's lobbying, not a decision — nobody's published the criteria the administration will use, and there's no announced timeline for when a host state actually gets picked. The substantive debate underneath the map fight is the one worth naming in each side's own terms: supporters frame the Academy as a way to train the next generation of space professionals, while critics worry it blurs the line between NASA's civilian mission and the military's. Both sides deserve their own words here, and neither published a knockout argument this week — what changed is simply that the fight over geography, over which state gets the jobs and funding that come with hosting a federal academy, is now live. Every state making a pitch this week is betting there's real money and real prestige in landing this.

That's the week — a five point nine two billion dollar crew contract, a nearly two billion dollar acquisition raise, and the Pentagon finally saying out loud what everyone assumed about weapons in orbit. If Rocket Lab's Mars orbiter protest gets a ruling from the GAO in the next few months, that tells us whether losing bidders on national-money missions actually have teeth. If today's recap helped you catch up on a busy week, send it to the one person in your life who still thinks space is just rockets. This has been Space Stakes, an AI-voiced podcast, created and built by a real human using today's cutting-edge technology. Nothing you heard on this show is financial advice. I'm Brian Lampert, and I'll catch you all tomorrow — take care!

I also host Quickly Quantum: a daily quantum computing briefing you don't need a physics degree to follow. The breakthroughs, the funding rounds, and how much substance is really under each claim. Find it wherever you get your podcasts.