On Thursday night around 9 p.m. Eastern, Blue Origin’s New Glenn exploded at Launch Complex 36 while undergoing a static fire test in advance of its first scheduled mission to carry Amazon Leo satellites.
The rocket was being prepared to launch 48 satellites but those satellites weren’t lost because they hadn’t yet been loaded. But they build the satellites quickly, the binding constraint on getting the satellite wifi service that Delta and JetBlue have committed to up and running is the launches. They can build 30 satellites per week, and there’s already a backlog of hundreds of them getting into space.
Blue Origin's New Glenn just blew up at LC-36 while attempting to Static Fire ahead of NG-4.https://t.co/tANS0dWyIH pic.twitter.com/PztxFoBqIw
— NSF – NASASpaceflight.com (@NASASpaceflight) May 29, 2026
We don’t yet know the extent of the launch pad damage. The transporter-erector and a lightning tower may not be salvageable. In all likelihood, New Glenn isn’t flying again this year. My interest is far more provincial: that this does to the timeline for Amazon Leo adoption inflight.
The explosion shouldn’t affect launches from other pads. But the incident removes a lot of the slack in getting to workable inflight wifi.
This angle is even crazier https://t.co/bDUuiafnTg pic.twitter.com/LuLG3frNw2
— Sawyer Merritt (@SawyerMerritt) May 29, 2026
The Amazon system needs around 600 – 700 satellites for limited beta. Starlink’s own minimally viable beta product was at around 900 satellites and that involved expected service interruptions with limited customers. So they need well over 1,000 satellites for meaningful service with consistent handoff between satellites. And full first gen service is going to mean over 3,000 satellites. Even there that’s only around a third of where Starlink is at.
Satellite count is not the same thing as capacity and coverage, but Amazon is already far behind and delays in getting to scale mean further delays for JetBlue (2027 roll out) and Delta (planned for 2028).
| Milestone | # Satellites | |
| Current production satellites before LA-07 | 302 | |
| Initial service-rollout phase | 578 | |
| Amazon’s July 30, 2026 projection | ~700 | |
| FCC 50% milestone | 1,618 | |
| Full first-gen constellation | 3,232 | |
| Versus Starlink | ~ 10,000 |
Amazon has 102 launches contracted across four providers: 18 Ariane 6, 24 New Glenn, 38 Vulcan Centaur, 9 Atlas V, and 13 Falcon 9. They plan more than 100 missions through the beginning of 2029, averaging around three launches a month at over 40 satellites per launch.
The New Glenn launches represented over 1,100 satellites. That capacity isn’t lost, but it’s delayed. We don’t yet know whether New Glenn returns early next year. And it’s not yet clear how many launches can be moved over to other providers. Losing New Glenn for a year could seriously affect the timeline. As it is they aren’t likely to meet their summer target for deployed Leo satellites and they’ve asked to move the FCC midpoint deadline to July 30, 2028.
JetBlue had planned Amazon Leo on 25% of its fleet next year. If New Glenn is out for 2027, that slows JetBlue’s rollout. Delta doesn’t begin installs until 2028, so that’s still plausible. But Amazon wasn’t going to have Starlink-style coverage (or use) then to begin with. Here’s a JetBlue plane potentially watching its install timeline blow up.
Video of the Aerial view of New Glenn exploding from Orlando Airport pic.twitter.com/WaYDFXYw1t
— Alex Quinn (@ItsAlexQuinn) May 29, 2026
There’s just less margin for error in hitting Delta’s timeline versus delay. And that’s the biggest issue: Starlink is available now. United and Southwest expect to have their full fleets Starlink-equipped next year before Delta even starts their Amazon installs already, if Amazon hits its timeline (and Delta only plans to move half its fleet to better wifi). The risk is that, in going for the Amazon as the cheaper option (because Delta can do a broader deal selling its customers to Amazon for shopping and content), Delta winds up at a strategic disadvantage for a longer period of years.
BOOM CRASH ️
Charli XCX. Well it was Boom, Clap. But that’s what came to my mind.
No worry though. For real. We’re moving technology at the speed of light (yes, lasers are used for satellites to communicate to each other now) and it’s crazy how much innivation happens in just a short six month period.
Damn, what a wonderful time to be alive! Pretty soon: 4K HD streams for every single passenger in every single seat in a tube above the earth flying at 600 mph — no matter if you’re sipping a pre-departure beverage or you’re stuffing yourself into 31K.
Don’t forget NASA and the USSR in the 50s and 60s were basically blowing up rockets literally once every few weeks. Vanguard, Atlas, Titan…. even the N1 Moon Rocket failed on ALL FOUR four launch attempts. I think they had like a 50-60% success rate?
We’re cool. All good in 2026. You just gotta look at history.
We have that now on United Mikey. I flew with Starlink yesterday morning streaming 4K. It is ‘the future’ at Georgia Klan Air maybe, but for other airlines the future is now.
SpaceX/Starlink: Fast, Vertical, Aggressive
SpaceX is widely regarded as the more nimble operator. SpaceX’s success stems from Musk’s willingness to bypass bureaucratic inertia, and the results speak for themselves in satellite connectivity. When the first Starlink satellites launched in May 2019, there were barely 2,000 active satellites in orbit total. Today there are more than 11,000 — the vast majority Starlink.
The core advantage is vertical integration. SpaceX’s 8,000+ satellites and reusable rockets enable rapid deployment in a way no competitor can match — they build their own rockets, launch them, and iterate on the satellite hardware simultaneously. This tight loop between manufacturing, launch, and operations lets them move fast in ways that a company dependent on outside launch providers simply can’t.
Amazon Leo: Large Company Navigating Its Own Scale
Amazon is the more bureaucratic of the two — though it’s actively trying to change that. CEO Andy Jassy has tried to reset Amazon’s corporate culture to operate like the “world’s largest startup,” setting internal targets to slash management layers and establishing a “no bureaucracy email alias” to identify ways the company can innovate faster. The fact that this is a stated initiative says something about the underlying problem.
The satellite program reflects that challenge. Amazon — despite its vast resources — has struggled to match SpaceX’s urgency in building out Leo. Amazon CEO Andy Jassy noted the technology is “on the verge of launching” in mid-2026, with Leo currently operating just 241 satellites against Starlink’s 9,000+. Amazon even had to hire SpaceX’s own Falcon 9 rockets to try to meet FCC deadlines — a striking sign of how far behind they are in launch capacity.
That said, Amazon’s slower pace isn’t pure bureaucracy — it’s also a reflection of being a massive, multi-industry conglomerate where satellite internet is one of many enormous bets. Amazon CEO Jassy has stated the company is ready to invest about $200 billion in 2026, primarily toward AI infrastructure, which means Leo is competing internally for resources and attention against AWS, AI chips, and logistics.
The Key Structural Difference
SpaceX is a rocket/satellite company. Every resource, every engineer, every launch is focused on that mission. Amazon Leo is a project within a trillion-dollar retailer/cloud company. That structural difference matters more than any cultural one — it shapes how fast decisions get made, how much risk leadership will absorb, and how quickly resources flow to problems.
Where Amazon does have a genuine edge is in enterprise integration: Amazon’s Prometheus chips transform Leo into a cloud-native platform, with AWS integration and telecom partnerships creating scalable monetization beyond consumer broadband. That’s a different kind of innovation — ecosystem depth rather than deployment speed.
Bottom line: For aviation WiFi specifically, Starlink’s speed-to-market advantage is decisive right now. Amazon Leo may close the gap over time, but it’s operating on a 2–3 year lag, and that’s largely a function of organizational structure, not just technology.
Delta didn’t use the premium space launcher for its premium company and look what happened. lol.
@Tim Dunn?
@Thing 1
When the troll stays in his cave, let him stay.
This rocket explosion feels like a metaphor for DL’s operational performance this summer.
(Succession, Season 1, Episode 10: Roman Roy rushes safety protocols to speed up a Waystar satellite launch in Japan to impress his father, Logan Roy. During Shiv’s wedding, he watches the launch on his phone, only to watch in horror as the rocket explodes shortly after liftoff.) Anyone else?