Is Elon Musk losing the space cellphone war?
The fiercest space race is not about getting back to the moon—it’s about allowing you to post a TikTok or watch Netflix on your phone anywhere around the globe, from the Atacama Salt Flats to the Khongor sand dunes in the Gobi Desert. To make this happen, two distinct design philosophies are at war, as companies build out the infrastructure needed to ensure every phone on the planet is permanently connected to the internet.
On one side is Elon Musk’s SpaceX/Starlink and the copycat companies that have followed in Starlink’s wake. Their approach is to invade space with tens of thousands of small satellites, creating a network of objects that blanket low Earth orbit. On the other side is a small Texas-based company called AST SpaceMobile, which believes it can provide better service with fewer than 100 gigantic satellites in space.
Both companies—along with Amazon and a handful of Chinese organizations—want to dominate worldwide wireless communications. The satellite constellation with the fastest service, widest coverage, best compatibility with 5G cellphones, and lowest operational costs will own how we communicate for years to come. Which approach prevails will have serious impact not only on the future of the internet but also the health of our planet.
A new space race era
Musk set off a new space race with his desire to rule low Earth orbit. SpaceX, which owns Starlink, launched its first satellite in 2019, providing broadband internet access to anyone with a large Starlink antenna and modem on the ground. Since then, it has put more than 9,000 satellites into orbit. The company projects it will eventually have a constellation of 34,000 satellites. After Starlink’s initial launch, competitors followed suit, including Jeff Bezos and his Project Kuiper—now called Amazon Leo—and the Chinese, whose plans include two large satellite constellations.
But there’s a fundamental problem with this mega-constellation design: Musk’s plan for space internet is a flawed, wasteful, and dangerous game of orbital Russian roulette.
Scientists worry that Starlink’s projected 34,000-satellite constellation will cause irreparable damage to the atmosphere. A large-scale constellation also dramatically increases the possibility of a space collision that could start a catastrophic chain reaction, destroying orbital networks that are crucial for our survival as a species.
Jonathan McDowell, an astrophysicist and spaceflight historian at the Harvard-Smithsonian Center for Astrophysics, has been documenting satellite launches in his newsletter, Jonathan’s Space Report. He believes there may be other, better ways to achieve global coverage via satellites—if we need to be doing it at all.
“I do personally have a preference for smaller numbers of larger satellites,” he tells Fast Company. “One of the reasons is the risk of space collisions. If you have 10 times as many satellites, you have 100 times as many close misses. So from that point of view alone, consolidating on a smaller number of satellites seems wiser.”
A more efficient alternative
That’s where Musk’s biggest competitor comes into play. AST SpaceMobile has developed a direct-to-cell technology that utilizes large satellites called BlueBirds. These machines use thousands of antennas to deliver broadband coverage directly to standard mobile phones, says the company’s president, Scott Wisniewski.
“This approach is remarkably efficient: We can achieve global coverage with approximately 90 satellites, not thousands or even tens of thousands required by other systems,” Wisniewski writes in an email. McDowell agrees that AST SpaceMobile’s approach is more efficient and less wasteful.
The key is its satellites’ size and sophistication. AST’s first generation of commercial satellite, the BlueBird 1-5, unfolds into a massive 693-square-foot array in space. Today, the company has five operational BlueBird 1-5 satellites in orbit, but its ambitions are much bigger. On December 24, 2025, AST launched the first of its next-generation satellites from India—called Block 2—and this one broke records. The BlueBird 6 has a surface of almost 2,400 square feet, making it the largest single satellite in low Earth orbit. The company plans to launch up to 60 more by the end of 2026.
“This large surface area is essential for gathering faint signals from standard, unmodified mobile phones on the ground,” Wisniewski explains. It is essentially a single, extremely powerful and sensitive cell tower in the sky, capable of serving a huge geographical area.
This design philosophy directly addresses the two greatest threats posed by the mega-constellation model. First, with only about 90 Block 2 satellites needed for global coverage, the sheer volume of material being launched and deorbited is orders of magnitude less than the tens of thousands planned by Starlink and others. With a 7- to 10-year lifespan, AST SpaceMobile’s satellites are designed to last longer than Starklink’s satellites, which have a lifespan of about 5 years. This combination of factors drastically reduces the potential for atmospheric pollution.
Additionally, a smaller number of satellites dramatically lowers the risk of orbital collisions. “Fewer satellites in orbit inherently reduces the probability of collisions and the creation of space debris, promoting a more sustainable orbital environment,” Wisniewski says. It is a solution built on precision engineering rather than brute numerical force, a testament to a different way of thinking about the problem.
As McDowell puts it, from a space traffic point of view, “Fewer, bigger satellites is probably better.” It is a design choice that prioritizes sustainability and risk mitigation.
A reckless, brute-force plan
The core idea behind Starlink’s direct-to-cell service is one of brute force. It is the digital equivalent of carpet-bombing: Saturate low Earth orbit with tens of thousands of relatively small, cheap, and disposable satellites. Each one acts like a tiny cell tower in the sky, talking to the phone in your pocket. Because they are in a low orbit, the lag is minimal, and the signal is strong enough for a standard phone. It’s a simple concept, but its elegance is deceptive. In reality, it has the elegance of a sledgehammer.
Starlink’s model relies on a constant cycle of replacement. The satellites are programmed to fall back to Earth after about five years, burning up on reentry. This is where the first major problem arises.
“When they burn up, they don’t just vanish,” McDowell explains. “They turn into dust, alumina dust, aluminum oxide particles. These particles are very good at destroying ozone.”
The long-term effect of depositing tons of this material into the upper atmosphere every single day is a terrifying unknown. We are, in effect, conducting an uncontrolled experiment on the protective layers of our own planet. McDowell notes that while a single rocket launch causes temporary, localized ozone damage, the continuous reentry of thousands of satellites creates a persistent, global problem that has never been studied on this scale.
SpaceX aggressively dismissed these concerns in 2021 in a legal battle with Viasat, a rival space internet service for home, business, and military use. Its legal defense directly attacked the scientific premise that burning satellites create harmful amounts of aluminum oxide. SpaceX has been ignoring warnings about potential ozone depletion since 2024.
However, the company has tried to address light pollution. When faced with an outcry from the astronomy community about its satellites’ brightness, it iterated on the design. First came DarkSat, an experimental coating that proved ineffective. Then came VisorSat, a deployable sunshade that blocked light from reflecting off the brightest parts of the satellite.
McDowell tells me that now SpaceX is using a dielectric mirror film that reflects less light back to Earth. “They have made a significant effort to reduce the brightness, and the newer Starlinks are substantially fainter than the early ones,” McDowell says. “But they are still bright enough to be a problem for the big survey telescopes like the Vera Rubin Observatory.”
These mitigation efforts, while commendable, address only one symptom of the problem—light pollution—and do nothing to solve the more fundamental issues of atmospheric pollution and orbital crowding. The problem is compounded by the fact that everyone is now copying the SpaceX model. Amazon’s Project Leo plans to launch more than 3,200 satellites.
Beijing and some Chinese companies are planning two separate mega-constellations, Guowang and G60 Starlink, totaling nearly 26,000 satellites. “We’re just at the beginning of this . . . so that gets very worrying because now it’s not just one company, it’s a whole bunch of companies,” McDowell warns. To add to his worries, just this week the Chinese government has applied for launch permits for 200,000 satellites.
To be clear, AST SpaceMobile’s approach is not without its own controversies. The sheer size of the company’s satellites makes them incredibly bright in the night sky, a significant source of frustration for ground-based astronomers. McDowell confirms that when it launched in 2022, AST’s prototype satellite, BlueWalker 3, became “one of the top 10 brightest objects in the night sky for a while.”
“It’s a serious issue, and we are working directly with the astronomy community to mitigate our impact,” Wisniewski says. The company is exploring solutions like anti-reflective coatings and operational adjustments to minimize the time its satellites are at maximum brightness. However, McDowell is not aware of anyone working with AST SpaceMobile, and the company didn’t provide any specifics.
According to McDowell, the size and brightness is a trade-off he believes is reasonable. “Although the BlueBirds are scary bright, there aren’t that many. So I kind of prefer that approach,” he says. “As long as they don’t turn around then and say, ‘Actually, we need 30,000 of these as well.’”
A game of orbital Russian roulette
Beyond the environmental concerns lies an even more immediate existential threat: Kessler Syndrome. Popularized by the movie Gravity, it is a scenario that keeps space experts like McDowell up at night. The theory, proposed by NASA scientist Donald Kessler in 1978, describes a domino effect where a collision between two objects in orbit creates a cloud of debris. Each piece of that debris then becomes a projectile that can cause another collision, creating even more debris, until low Earth orbit becomes an impassable minefield of hypervelocity shrapnel.
“The more satellites you have, the more the chance of a collision,” McDowell states plainly. “And the problem is once you have the first collision, the debris from that is now threatening all the other satellites.”
SpaceX has engineered a highly automated collision avoidance system for Starlink, and McDowell acknowledges its sophistication. The company’s satellites constantly monitor their trajectories and can autonomously fire their thrusters to dodge potential impacts. “They do thousands of maneuvers a month,” he says, which is a testament to both the system’s capability and the terrifyingly crowded environment it operates in. In total, Starlink satellites have performed 50,000 evasive maneuvers since 2019.
But while SpaceX claims that its satellites are 100% safe, the facts tell us that they are not foolproof. “Even with a 99% success rate for deorbiting, a 1% failure rate on a 30,000-satellite constellation means you’re adding 300 dead, multi-hundred-kilogram satellites to orbit every five years,” McDowell says. That’s 300 uncontrollable bullets waiting to start the Kessler Syndrome.
A catastrophic chain reaction could, in a matter of hours or days, wipe out the essential satellite networks that underpin modern civilization. This isn’t just about losing your GPS navigation on the way to a new restaurant. It’s about the collapse of global finance, weather forecasting, communications, and critical military and disaster-response systems. We are talking about a technological regression of decades, a scenario McDowell finds increasingly plausible as more mega-constellations are launched. It’s a high-stakes gamble with civilization’s essential infrastructure.
There’s also a direct-hit danger for people on the ground. A few Starlink satellites have already failed in orbit, becoming uncontrollable space junk that fell back to Earth. There’s at least one report of a piece of a satellite hitting a building in Canada. The latest reported incident took place on December 17, 2025, when a Starlink satellite experienced an anomaly, losing communication and causing a propulsion tank vent, rapid orbital decay, and the release of debris in low Earth orbit. In a 2023 report to congress, the Federal Aviation Administration said there’s a real risk of falling Starlink debris injuring or killing someone by 2035.
Space junk is also a problem for rockets. In early November, three taikonauts returned after being stranded on China’s Tiangong space station for nine days. They couldn’t use the spaceship that was going to take them to Earth—the Chengdou-20—because it had been struck by orbital debris. The China Manned Space Agency said its astronauts found “tiny cracks” in a small window of their Shenzhou-20 spacecraft.
The hit was not fatal, but things could have gone very wrong. The Chinese, however, seem undeterred. Beijing will be launching hundreds of thousands of satellites that mirror Starlink’s design, contributing to the problem and increasing the risk to themselves and everyone else.
Why AST SpaceMobile could win
Right now, Starlink doesn’t provide direct-to-cell broadband; instead, it provides only text and limited data connections. This low-speed connectivity requires a line of sight with the satellite, as SpaceX states on its site. Starlink, despite its leading market position in the internet satellite business, is still playing catch-up on the direct-to-cell front, and it may never be able to close the gap with AST.
Musk’s company has two big strikes against it. First is the hardware in orbit. To provide broadband to phones it needs a next-generation Starlink V3 satellite, which doesn’t exist yet. SpaceX has no reliable way to launch it, anyway. At an estimated 4,400 pounds, the V3 satellite is too big and heavy for the workhorse Falcon 9 rocket to deploy in economically viable numbers.
The entire business model for Starlink V3 hinges on the success of Starship, Musk’s next-generation, super-heavy-lift launch vehicle. But Starship remains in development, having yet to achieve the consistent operational launch cadence required to deploy and maintain a constellation of thousands of V3 satellites.
But even if SpaceX manages to finish Starship and Starlink V3 satellites on time, there’s a second, even bigger hardware problem: The broadband connectivity won’t work unless the cellphone has a special modem chip. Yes, my space cadets, you will need to buy a new phone to enjoy Starlink connectivity, while AST works with any current, unmodified phone. None of these new Starlink-enabled phones exist or have been announced yet. According to SpaceX CEO Gwynne Shotwell, the company is now working on this chip.
“We’re working with chip manufacturers to get the proper chips in phones,” she told the audience at World Space Business Week in Paris back in September. Expecting phone manufacturers to incorporate Starlink’s proprietary modem in their phones feels like a tall order. Especially when manufacturers like Apple have their own direct cell-to-space plans. It seems unlikely that Tim Cook will tie his company’s crown jewel to Musk’s whims. Or make the phones even more expensive.
However, that doesn’t matter, because even if Musk had 15,000 V3 satellites and the Starships ready to launch, the problem will remain the same: You will need phones with Starlink modems in them for broadband. And the line-of-sight problem will persist. The broadband works only when the phone can “look” at the satellites in the sky.
This is why Musk’s promised direct-to-cell broadband timeline is speculative. In fact, while he said it would be ready in 2026, according to SpaceX, testing of the first phones equipped with Starlink chips is scheduled for this year, with an aim to complete its V3 direct-to-cell satellite constellation in 2027. It is a promise built on a promise, a technological if dependent on a logistical when. Meanwhile, AST SpaceMobile is preparing to launch new operational satellites on existing rockets.
AST SpaceMobile has already proven its technology works, with six working satellites now transmitting at typical 5G speeds directly to regular phones. This doesn’t mean that it has a guaranteed win against Starlink or any of its competitors. While it has the capital to execute its plan—with the backing of investors like AT&T, Alphabet, Rakuten, Vanguard, BlackRock, and Mexican magnate Carlos Slim, who owns the largest telecom operator in Latin America—and superior technology, it needs to execute dozens of launches.
This confidence in the technology explains why the stock has skyrocketed 333% in a year, but the doubts about potential execution problems also explain why the stock experiences wild swings. With every news of a launch or a delay of a launch, the stock can swing 10% or 20% up and down. That’s why the market treats AST as a high-risk, high-reward battleground. It can be a trillion-dollar business or explode on the launchpad if the company doesn’t put all those satellites up in 2026.
The coup de grâce
This technological and philosophical divergence has not gone unnoticed by Elon Musk. Seeing a direct threat to his ambitions, he has engaged in a campaign to undermine AST SpaceMobile with baseless accusations, claiming that its satellites are a danger in low Earth orbit because of their size.
At the same time, Musk is battling his own problems in low Earth orbit. China has already denounced two near misses with Starlink satellites that triggered its space station to perform emergency evasive maneuvers. On January 2, SpaceX was forced to move 4,000 satellites to a lower orbit after new research by Chinese scientists highlighted the company’s recklessness and the very real risk of collision.
Moreover, Musk desperately tried to stop the Federal Communications Commission from granting AST access to the necessary spectrum—the range of radio frequencies it needs for its satellites to connect with cellphones on Earth—claiming it would be “catastrophic” for his service because the powerful signals from AST’s large satellites could interfere with Starlink’s user terminals. In response, AST SpaceMobile asserts that its system is designed to coexist with other networks and operates fully within the internationally agreed-upon limits established to prevent such interference. The FCC agreed and allowed AST to use the spectrum.
And that move, if every logistical aspect executes to plan, gives AST an absolute slam dunk against Musk.
Think of the radio spectrum as a giant highway in the sky with a limited number of lanes. Carrying data back and forth, AST’s trucks have the rights to travel through a huge number of these lanes thanks to partners like AT&T and Verizon—roughly 35 MHz of what the industry calls the “golden low-band spectrum.” It also acquired an extra 45 Mhz low-band spectrum from a bankrupted communications company called Ligado. That’s a massive 80 MHz.
And remember the company’s patented magic sauce we mentioned earlier? That’s AST’s secret weapon to make this highway work: a chip that glues the different radio bands together into one massive pipe, capable of delivering peak speeds of 120 megabits per second to phones (comparable to your typical 5G connection).
SpaceX and its partner T-Mobile have very few lanes available right now: only 5 MHz. That’s like comparing an 80-lane superhighway to a 5-lane street. To try to fix that, Musk has spent $17 billion to acquire 50 more lanes: 50 MHz of S-band spectrum from another bankrupt communications company, EchoStar. The problem is that physics dictates that higher-frequency radio waves, those that SpaceX is operating on, do not penetrate solid objects as effectively as lower-frequency waves. That’s why Starlink’s space-to-cell service will require line of sight to work.
Meanwhile, AST claims its system will work indoors and outdoors, penetrating buildings in a similar way to a regular cell signal. Wisniewski claims that a phone will connect through “one wall” and work through your car’s roof because of two factors: It uses a low-frequency radio connection, and its satellites are big enough to “listen” and “talk” to the phones on the ground, even behind obstacles.
This scenario has yet to be proven by AST or a third party. Only line-of-sight broadband with regular phones has been tested successfully. However, if it works inside cars and buildings like Wisniewski claims, the user experience will be seamless. A phone will have service where it didn’t before, delivered through an existing provider like AT&T or Vodafone.
Should we really do this?
But what if we’re looking at this cell-to-space race the wrong way?
“I hear from friends who go hiking in faraway places, look up at the sky, and say, ‘Wow, you just never see an empty sky anymore,’” McDowell tells me with a hint of sadness and worry. Musk, Bezos, Wisniewski, the Europeans, and the Chinese would argue that we need ubiquitous cheap internet everywhere in the world, for civilian and military applications. Sure, there’s a lot of money to be made and there’s a genuine need to serve communities in remote places without having to invest in ground infrastructure. But do we really need to stream TikTok from space?
“I don’t necessarily have a position on that,” McDowell says. “My position is that even if that’s the case, it shouldn’t be just the U.S. that decides that. It should be decided by all of the countries in the world, because they’re all affected whether they’re space powers or not.”
He’s right. And if we all decide that we do need this, we should also all agree on the best solution to minimize the impact on humanity. The smarter solution. The most technologically advanced.