SpaceX launches first orbital Starship attempt as UK stands up a space squadron

The launch window opened at 8:15 a.m. ET, or 7:15 a.m. local time, according to SpaceX's website, as cited by CNBC. The Federal Aviation Administration told the network by email that it issued a "license authorization for the SpaceX Starship Super Heavy Flight 14 launch and reentry operations" on 26 September.

None of the 13 earlier test flights were meant to reach orbit. Flight 14 is supposed to circle the planet and, on the same mission, deploy 26 of SpaceX's new Starlink V3 satellites.

What is actually new on this flight

The V3 satellites are built at a SpaceX facility in Redmond, Washington. According to CNBC, they carry solar arrays that generate twice as much power as earlier generations. They are also physically larger, and they are designed to transmit data faster to Starlink subscribers.

Scale is why the company cares. SpaceX operates around 11,000 active satellites in orbit, against roughly 650 for competitor Eutelsat OneWeb, CNBC reports. On an August earnings call, Elon Musk said: "It's not out of the question that at some point, Starlink will deliver a majority of the world's internet, at least in countries where we're allowed to operate, which is the vast majority of countries."

Starship is the vehicle for that expansion. SpaceX has designed the Super Heavy stack to be fully reusable and to lift more cargo for less cost than any prior spacecraft. The company is also preparing the rocket for a major NASA test flight next year and intends to use it to return US astronauts to the lunar surface.

Where the science result comes in

The scientific value of Flight 14 is narrow but testable: getting a fully reusable upper stage through reentry and to a controlled splashdown, then doing it again. Reusability is what turns a rocket into infrastructure, and infrastructure is what satellite constellations and science payloads are priced against.

Orbital insertion itself is not a science result. The result is the data set from a vehicle that reached orbit and came back in one piece. That distinction matters because SpaceX has been here before without crossing the line. The company went public in June in the largest IPO on record and is now valued at about $2 trillion, according to CNBC, a valuation that assumes Starship stops being a test programme.

There is a second payload class that gets less attention. SpaceX COO Gwynne Shotwell said during the All-In Summit earlier this month that the company plans to use Starship to put "supercompute in space," launching "AI compute satellites" into orbit in 2027.

Whether that plan survives contact with orbital mechanics is an open question. But it explains why a Starlink deployment and an orbital test are being flown together rather than sequentially.

Britain's answer: fight satellites with jamming, not missiles

Two days before the Starship window, the Royal Air Force stood up a new unit. Air Chief Marshal Sir Harvey Smyth announced No. III Space Effects Squadron at an inaugural UK Space Power Conference in London, according to The Register.

Smyth said the squadron will safeguard the benefits space gives to UK armed forces and be capable of denying those benefits to adversaries. The RAF's headquarters at High Wycombe transitioned to become Air and Space Command a few weeks ago, he added.

The existing No. I Space Operations Squadron and No. II Space Warning Squadron give the RAF the ability to watch and warn. No. III is meant to act.

How is less clear. The UK has limited launch facilities and, as far as is publicly known, no anti-satellite missile systems. The government says the capability will be delivered using "advanced technology," including electronic warfare, to "disrupt, degrade and deny hostile actions." The Register asked the Ministry of Defence for specifics; a spokesperson said not much more could be said because the information is classified.

The MoD does say the squadron will support both offensive and defensive space operations and will expand as new capabilities enter service.

The policy is not classified. The UK has formally ruled out destructive anti-satellite missile testing and works through the UN Committee on the Peaceful Uses of Outer Space. That implies ground-based jamming rather than kinetic weapons, which The Register notes is consistent with the wording.

Four countries are known to have tested anti-satellite missiles: the US, Russia, China and India. The most recent incident came in 2021, when Russia destroyed one of its own defunct orbiters and endangered the International Space Station. The same year, the ISS manoeuvred to avoid debris created by China's earlier destruction of one of its own satellites. In 2022 the UN passed a US-led resolution calling on countries to commit not to conduct similar tests.

That has not stopped interest in non-kinetic options. China has looked at lasers and high-power microwaves to disable orbital assets, according to The Register, and the US Space Force has claimed to have observed Chinese satellites practising "dogfight" manoeuvres in orbit. In 2025, The Register reported that the MoD was working with the UK Space Agency on technology to detect when lasers are used to dazzle satellites and disrupt communications.

The terrestrial bottleneck nobody solved

The same week, a solar project in Illinois showed how constrained the ground side of the space and energy economy has become. CleanTechnica reported on 27 September that Cultivate Power's Bowes Solar array, an 11-megawatt (DC) installation, fits onto a 22-acre site in Elgin wedged between two residential subdivisions, high-voltage power lines, train tracks and a wetland.

Conventional community solar needs 5 to 7 acres per MWp, according to Planted Solar, the racking supplier. By that rule the Bowes array could not fit. Planted's software-enabled racking requires about 2 acres per MWp and can fill irregular spaces that conventional layouts leave empty. Cultivate split the installation into two arrays to handle setbacks and avoid the wetland.

Chronicle Media, cited by CleanTechnica, reported that the project will cut rates for 580 low- and moderate-income households and that subscribers will save $13.9 million on Commonwealth Edison bills over 40 years. Property tax revenues are expected to total $2.1 million over the same period.

"Developing solar on constrained parcels is a growing focus for the industry," Cultivate told CleanTechnica in an email. "It's becoming more important as competitive land use drives solar onto smaller and smaller parcels."

The money question underneath the AI satellites

Shotwell's 2027 orbital compute plan belongs to a debate that The Decoder framed on 26 September with a single question: is revenue at AI labs growing fast enough to pay for the data centre buildout?

Azeem Azhar, founder of the research group Exponential View, offered one data point on a podcast with Nicholas Thompson of The Atlantic. Speaking to about 160 IT vice presidents in Las Vegas, Azhar asked who could point to measurable AI results. Two-thirds stayed standing, more than he expected. He then asked who had results good enough to interrupt the CEO's summer vacation. About eight people remained on their feet.

Azhar's reading is that companies are making progress, slowly. The Decoder notes that a Boston Consulting Group survey found about 70 percent of CEOs worldwide say AI success matters to how they are perceived, which gives them an incentive to paint a rosier picture. Azhar calls the situation "finely balanced" and says he does not have a simple answer.

That is the honest state of play. Starship may reach orbit on Monday. Whether the economics that pay for the next 11,000 satellites, or the compute satellites after them, hold up is a separate test with no launch window.