SpaceX
SpaceX’s first dedicated Starlink launch announced as mass production begins
SpaceX has announced a launch target of May 2019 for the first batch of operational Starlink satellites in a sign that the proposed internet satellite constellation has reached a major milestone, effectively transitioning from pure research and development to serious manufacturing.
R&D will continue as SpaceX Starlink engineers work to implement the true final design of the first several hundred or thousand spacecraft, but a significant amount of the team’s work will now be centered on producing as many Starlink satellites as possible, as quickly as possible. With anywhere from 4400 to nearly 12,000 satellites needed to complete the three major proposed phases of Starlink, SpaceX will have to build and launch a minimum of ~2200 satellites in the next five years, averaging 37 high-performance, low-cost spacecraft built and launched every month for the next 60 months.
A shift in the Stars
Despite the major challenges ahead of SpaceX, things seem to be going quite smoothly with the current mix of manufacturing and development. As previously reported on Teslarati, SpaceX CEO Elon Musk forced the Starlink group through a painful reorganization in the summer of 2018, challenging the remaining leaders and their team to launch the first batch of operational Starlink satellites no later than June 2019. As a consequence, a sort of compromise had to be reached where one additional group of quasi-prototype satellites would be launched before settling on a truly final design for serious mass-production.
According to SpaceX filings with the FCC, the first group of operational satellites – potentially anywhere from 75 to 1000 or more – will rely on just one band (“Ku”) for communications instead of the nominal two (“Ku” and “Ka”), a change that SpaceX says will significantly simplify the first spacecraft. By simplifying them, SpaceX believes it can expedite Starlink’s initial deployment without losing a great deal of performance or interfering with constellations from competitors like OneWeb.
Somewhere along the line, SpaceX would iteratively improve each subsequent ‘generation’ of Starlink satellites until they reached the nominal performance characteristics outlined in the company’s original constellation application. Knowing SpaceX, improvements would continue for as long as lessons continued to be learned from operating hundreds and eventually thousands of orbital spacecraft.
As one concrete example, recent SpaceX FCC documents stated that the first 75 Starlink spacecraft would feature a less-optimized reentry design, meaning that a select few components will not entirely burn up during reentry, creating debris that poses a
While the FCC has yet to grant SpaceX’s requested modifications, the other major goal is to reduce the operating orbit of the first phase of 1584 satellites to 550 km (340 mi), a change that SpaceX says will drastically reduce the potential lifespan of any orbital debris in the unlikely event of their creation. A lower altitude also places a major cushion between SpaceX’s first ~1500 satellites and the orbits of several other planned constellations, including OneWeb and Telesat.
Hello, Production Hell, my old friend
Meanwhile, SpaceX’s Starlink program has begun the often painful steps of transitioning from a venture primarily focused on research and development to one focused mainly on building production lines and supply chains and manufacturing hardware. SpaceX’s Starlink facilities are currently housed in three nearby buildings located in Redmond, Washington, likely offering approximately 150,000 square feet (14,000 m^2) for a mix of office, development, and production spaces. At least one of the three non-office buildings could potentially become dedicated to production while one building – approximately 40,000 ft^2 (~3500 m^2) – has already been completely transformed into a prototype of a Starlink satellite production line, supporting manufacturing for first several dozen quasi-prototype spacecraft. For reference,
Mass-producing spacecraft at the scale needed to build even half of those needed for the first phase of ~4400 Starlink satellites will be a feat unprecedented in the history of the space industry. Barring FCC exemptions (possible but unlikely), SpaceX needs to launch ~2200 Starlink satellites between now and April 2024. To complete the first phase, the final number of satellites rises to ~4400. Adding on a proposed constellation of very low Earth orbit (
Perhaps SpaceX will be able to garner invaluable insight from the lessons its sister company learned during Model 3’s torturous “production hell”, in which the car company had to grow its production volume by almost a magnitude as quickly as possible. Ironically, it may even be the case that SpaceX has the easier task relative to Tesla.
Check out Teslarati’s Marketplace! We offer Tesla accessories, including for the Tesla Cybertruck and Tesla Model 3.
News
SpaceX tells the FCC that Starship Flight 14 is going to orbit
SpaceX filed with the FCC for Starship Flight 14, its first true orbital launch attempt.
SpaceX has asked the Federal Communications Commission for permission to fly Starlink terminals during Starship’s fourteenth flight test, and the filing lays out a genuine trip to orbit, something the program has never attempted.
Every Starship flight so far, including Flight 13’s successful splashdown in the Indian Ocean on July 24, has flown a suborbital arc that ends with the ship reentering the atmosphere within the same hour it launches. The FCC paperwork describes a mission profile built around an actual orbital insertion instead.
The payload is the other half of the story. Flight 13 carried 20 production Starlink V3 satellites, but because that mission never reached orbit, the satellites reentered along with the ship rather than joining the constellation, something Teslarati covered in detail after SpaceX released footage shot from one of those satellites as it drifted away from Starship in space. Flight 14 is designed to close that gap. If the orbital insertion holds, the roughly 20 V3 satellites onboard would separate into an operational orbit and could eventually go into service, each one rated for about 1 terabit per second of downlink capacity by SpaceX’s own account.
SpaceX announces new Starbase for ‘thousands of Starship launches annually’
Elon Musk first flagged the orbital attempt during SpaceX’s August 4 earnings call, the company’s first as a public entity following its June IPO under the ticker SPCX. He also floated catching the ship with the Starbase tower on the same flight, an idea he walked back on August 20, saying the catch attempt would more likely come “in a few months,” as Teslarati reported at the time. Flight 14 will instead target a splashdown for the ship in the Indian Ocean, the same recovery method used since Flight 12.
Hardware has been catching up to the ambition. Booster 21 completed a full 33-engine static fire on August 28, and Ship 41 finished its own six-engine test the week before. An airspace briefing circulated to pilots on August 20 listed September 15 as the target date, later than the end of August window Musk mentioned on the earnings call, though SpaceX has not confirmed a launch date publicly and Starship schedules routinely slip while hardware and FAA paperwork line up.
The FCC filing itself does not guarantee a launch date. It covers communications authority, and not flight readiness, considering SpaceX still needs Ship 41 fully stacked and cleared by the FAA before Flight 14 can fly. But the filing is a real marker of intent and it puts a specific regulatory process behind what had so far only been Musk’s word on the earnings call.
Elon Musk recently restated a fact that still defines SpaceX’s origin story: if Falcon 1’s fourth launch had failed, the company would not exist. The comment answered a reminder that after three consecutive losses, SpaceX had money for only one more attempt.
On X, Peter Diamandis said that the present-day acknowledgement of SpaceX’s success does not discount the rough start the company had. “Almost nobody remembers that Elon’s first rocket failed three times, and there was money for exactly only one more attempt.”
Musk said, “If the 4th launch had failed, SpaceX would not exist.”
If the 4th launch had failed, SpaceX would not exist
— Elon Musk (@elonmusk) August 30, 2026
In late 2008, the firm was nearly out of cash. Another failure would have ended payroll, closed the Hawthorne factory, and left the Falcon 9 and Dragon programs as unfinished drawings.
The first flight lifted off from Omelek Island on 24 March 2006. Thirty-three seconds later, a corroded aluminum fitting on a fuel line leaked. Kerosene ignited around the Merlin engine, control was lost, and the vehicle came apart. The small DARPA payload, FalconSAT-2, survived the short flight only to land on a storage shed near the pad. Investigators later traced the fitting to a materials mix-up that should never have reached the rocket.
Flight 2, on 21 March 2007, looked far better at first. The first stage burned cleanly and handed off to the Kestrel-powered upper stage. The vehicle crossed 100 kilometers and reached a peak of about 289 kilometers. Then propellant slosh in the second-stage tank started a circular coning motion that grew until the engine shut down. Telemetry faded as the stage tumbled, and SpaceX had reached space but not orbit. Over the next year, the team redesigned everything from the ground up, including tanks, baffles, and the new regeneratively cooled Merlin 1C.
That engine flew on Flight 3 on 2 August 2008. The first stage performed almost perfectly and reached 217 kilometers. After main-engine cutoff, leftover fuel in the cooling channels produced a faint residual thrust, roughly 10 pounds per square inch of chamber pressure. On a Texas test stand, the effect was invisible beneath ambient air pressure. In vacuum it was enough to push the spent first stage back into the second stage after separation. The stages collided, the upper stage spun, and the mission was lost. Musk later said a slightly longer delay before staging would have saved the flight.
Six weeks later, the team assembled Flight 4 from remaining parts and flew it on 28 September 2008 at 23:15 UTC. The payload was Ratsat, a 165-kilogram aluminum mass simulator built in-house. Staging was delayed so residual thrust could decay. The Kestrel ignited, the fairing split away, and nine and a half minutes after liftoff the vehicle was in orbit. After a coast, the second stage restarted, settling into a 621-by-643-kilometer path at 9.35 degrees inclination. Falcon 1 became the first privately developed liquid-fueled rocket to reach Earth orbit. Musk called the insertion “middle of the bull’s-eye.”
SpaceX restores a Falcon 1 rocket for 10th anniversary of first launch success
That success unlocked NASA’s Commercial Resupply Services award later that year. Without it, there would have been no Falcon 9, no reusable first stages, and no Dragon cargo or crew flights to the International Space Station. Launch prices would have remained far higher. Starlink’s constellation would not exist; broadband from low Earth orbit would still be a paper concept.
Ride-share markets, high launch cadence, and the current pace of lunar and Mars hardware would be years behind. Communications, Earth observation, and the cost of putting anything into space would look more like the 2000s than the 2020s.
One extra second of residual thrust in August 2008 would have written a different decade.
Elon Musk
OpenAI cites distrust of SpaceX in decision to drop Cursor partnership
OpenAI will cut SpaceX-owned Cursor’s model access in November, citing Musk’s history of broken contracts.
OpenAI, the company behind ChatGPT, announced late Friday that it is ending its partnership with Cursor, cutting off the coding tool’s access to its models on November 12. The move comes two weeks after SpaceX completed its $60 billion acquisition of Cursor’s parent company, Anysphere, folding the popular AI coding assistant into Elon Musk’s growing SpaceXAI division.
In a post on its website, OpenAI said the decision came down to trust, not technology. “We cannot be confident that SpaceX will use our technology within our terms of service, based on our experience with Elon Musk’s companies violating contracts,” the company wrote. OpenAI pointed to two specific incidents: X, now part of SpaceX, allegedly breaking the terms of an existing OpenAI contract after Musk bought Twitter.
That lawsuit is the backdrop for all of this. Musk cofounded OpenAI in 2015, left the board in 2018, and sued Sam Altman and Greg Brockman in 2024, arguing they abandoned the company’s nonprofit mission for profit. A federal jury sided with OpenAI in May, finding Musk waited too long to sue rather than ruling on the merits of his claims. Musk said at the time he would appeal to the Ninth Circuit, calling the outcome a “calendar technicality” rather than a real judgment.
SpaceX’s interest in Cursor predates that verdict by weeks. The company first struck a deal with Cursor in April, securing an option to acquire it for $60 billion or pay $10 billion for joint development work instead. As Teslarati reported at the time, the logic was straightforward: Cursor was paying retail prices to Anthropic and OpenAI, two of its most direct competitors, every time a developer used its product, while SpaceX had idle capacity on its Colossus supercomputer, roughly the equivalent of a million Nvidia H100 GPUs, that Cursor could use to train its own models instead. SpaceX exercised the option in June, days after its own IPO, and the deal closed in mid-August.
Once it closed, Musk moved fast. On an all-hands call with more than 1,000 Cursor employees, he reportedly told staff that SpaceXAI’s Grok was playing catchup in the AI race, unlike Tesla and SpaceX in their own markets, and singled out Anthropic as the company to catch. Cursor CEO Michael Truell now reports directly to Musk inside SpaceXAI.
Losing OpenAI’s models leaves Cursor leaning harder on Anthropic’s Claude, which has its own compute agreement with SpaceX, and on Cursor’s in-house Composer model, the one SpaceX’s compute was supposed to accelerate in the first place. OpenAI framed the November deadline as maximum notice under its contract, and said it wants to “go above and beyond” to help developers through the transition. Whether Anthropic makes the same call is now the open question in AI coding.
SpaceX tells the FCC that Starship Flight 14 is going to orbit
Tesla Cybercab Event: what to expect from Austin
