We predicted the X Phone months ago—and now Elon’s confirmed it. With AI inference on every device and 3B customers by 2030. This won’t just be a phone. It is the end of telecom as we know it.
Totally worldchanging within 5 years.
The $17B EchoStar deal involves AWS-4 spectrum (H-block/S-band), enabling high-bandwidth satellite-to-phone globally.
Musk hinted at a Starlink phone and partner company phones at the All in Summit 2025. Hardware changes like chipset modifications for new frequencies are needed and will be ready in 2 years. This will allow high-bandwidth video streaming anywhere with those phones working indoors in normal homes but not thick metal roofs.
Everyone in the audience raised hands to say they want a Starlink phone.
EchoStar, an old-school geosynchronous satellite company. Has lost half its customers to Starlink’s low-Earth orbit (LEO) constellation (8,000+ satellites). They are roadkill for SpaceX.
EchoStar failed to utilize the spectrum (FCC requires half-system launch within timelines). They sold it preemptively. SpaceX got $8.5B in EchoStar stock, boosting their value 20%.
In India and elsewhere, spectrum is a public resource. Underutilization leads to reallocation because the government auctions it to companies that will pay more and do more with it. If SpaceX is 100X more effective then the spectrum will get auctioned to SpaceX as the governments will get way more value by having the spectrum in competent hands.
My SpaceX Projections are
Now 7 million and 9-10 million customers by end of 2025. $5 billion in profit.
2026— 28M customers → $60B revenue $48B profit (high margins as expenses scale minimally). At 40x P/E (like Nvidia), valuation $1.6–2T.
By 2028 SpaceX will have 100M high speed internet users for $400B/year profit and a $16 trillion valuation. They will be the world’s largest company.
This is just from dishes/high-speed internet, not mobiles.
Mobile potential: India alone (300M+ users), plus Europe, Africa, etc.
Starlink could reach 3 billion customers by 2030.
All of that cash, profits and valuation will make it very easy to acquire more spectrum and more companies.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
I did a quick analysis (with GROK) on the problem with the AWS-4 spectrum overlapping the existing terrestrial 4G/5G networks. It turns out there will be problems in 20 – 30 countries requiring individual regulatory coordination.
In these countries, there are ~1.4 billion mobile subscriptions. Interestingly enough, there a many more subscriptions than people. People have more than 1 subscription each on average.
This is a market that can only be addressed by cooperating with existing operators.
The FCC auctions spectrum, but not LEO space. In fact, there is currently no national, or what’s needed more, international body to allocate LEO space, which is finite too, just like spectrum. Even though it’s 3D and large, 40,000 satellites as SpaceX intends to deploy, at 300km above Earth, yields ~118km between them, according to Supergrok (see its detailed formula from any earlier comment on this subject or repeat below). That’s narrow enough to be a problem for space flight between them, or other satellites for a rival telecom system, Bezos wants to do, or another country like China. And no one gave Starlink international permission to monopolize 300km LEO. And this is before any Kessler syndrome, almost inevitable since there’ll be competitor satellites and other debris.
Grok says:
“Calculation of Average Distance Between Starlink Satellites at 300 km AltitudeTo calculate the average distance between 40,000 Starlink satellites deployed uniformly in a spherical shell at an altitude of 300 km, we can approximate the satellites as being evenly distributed over the surface of a sphere with radius equal to Earth’s radius plus the altitude. This is a common method for estimating average separation in satellite constellations, treating it like a grid spacing on the sphere’s surface.Step-by-Step Explanation:Determine the orbital radius:
Use Earth’s average radius R=6371R = 6371R = 6371
km. Add the altitude h=300h = 300h = 300
km:
Orbital radius r=R+h=6371+300=6671r = R + h = 6371 + 300 = 6671r = R + h = 6371 + 300 = 6671
km.
Calculate the surface area of the sphere:
The surface area ( A ) of a sphere is given by:
A=4πr2=4π(6671)2≈559,231,654A = 4 \pi r^2 = 4 \pi (6671)^2 \approx 559,231,654A = 4 \pi r^2 = 4 \pi (6671)^2 \approx 559,231,654
km².
Find the area per satellite:
With N=40,000N = 40,000N = 40,000
satellites:
Area per satellite = A/N≈559,231,654/40,000≈13,981A / N \approx 559,231,654 / 40,000 \approx 13,981A / N \approx 559,231,654 / 40,000 \approx 13,981
km².
Approximate the average distance:
Assuming a roughly grid-like distribution, the average distance ( d ) (or typical separation) between satellites is the square root of the area per satellite:
d=A/N≈13,981≈118d = \sqrt{A / N} \approx \sqrt{13,981} \approx 118d = \sqrt{A / N} \approx \sqrt{13,981} \approx 118
km.
This is a high-level approximation and assumes ideal uniform coverage. In reality, Starlink satellites are arranged in specific orbital planes and inclinations, which could slightly adjust the actual average nearest-neighbor distances depending on latitude and orbital parameters. For a random (Poisson) distribution, the expected nearest-neighbor distance would be smaller, around 59 km, but constellations like Starlink aim for more even spacing to optimize coverage, making the grid approximation more appropriate.At a lower altitude like 300 km (compared to the current ~550 km orbits), the orbital sphere is smaller, leading to closer average spacing for the same number of satellites, which aligns with goals for better per-satellite coverage and lower latency.”
Average current global high speed internet bills range from a high max around $100 a month down, $50 or so in Europe, $40 S.America & Asia, $20 Eastern Europe,
If Starlink somehow managed to get a MUCH higher average fee than users pay now at $100 a month global and had 100M users it would take in $100 x 12 months x 100M users = $120B/yr gross revenue. Where does the $400B/yr profit come from?
That’s still very good growth but I want to understand your projections.