The internet has a difficult job.

It has to move ridiculous amounts of data, cheaply, reliably, and with predictable delay. Today that job is mostly done by cables.

But could “all internet” eventually move to space? In practice, no. But a meaningful slice of how we reach the internet will, and the slice will keep growing in the places that matter for disruption.

Space is winning the access layer, not the backbone

When people say “internet from space,” they usually mean the last mile problem.

Getting broadband to a farmhouse, a ship, a plane, a disaster zone, or a moving convoy is expensive with towers and fiber, and sometimes impossible on any useful timeline.

Low Earth orbit constellations fix that by turning “build local infrastructure” into “ship a terminal.”

That is why Starlink has pushed past thousands of satellites and keeps expanding service modes like maritime, aviation, and direct-to-cell. Recent reporting puts Starlink at over 9,500 active satellites and still launching.

This is the disruption. Space is becoming a default option for coverage, mobility, and continuity. It is also turning connectivity into a software-defined service where capacity can be reallocated dynamically across regions, industries, and customers.

But that is not the same thing as moving the internet’s core into orbit.

The core of the internet is a capacity game, and cables still win

International data traffic is dominated by submarine cables.

The ITU says subsea cables carry about 99 percent of the world’s internet traffic.

That number is not an accident. Fiber has three structural advantages that are hard to beat.

First, raw throughput. A single modern subsea cable system can carry huge volumes using many fiber pairs and dense wavelength multiplexing. Satellites have to share limited radio spectrum, deal with interference constraints, and cope with weather effects on some bands.

Second, economics. Once a cable is in the water, the marginal cost per delivered bit is extremely low. Satellites are manufactured hardware with finite lifetimes, constant replenishment, and launch costs. Even if launch gets cheaper, you are still replacing a global fleet on a cadence that cables do not require.

Third, topology. Most traffic is not “random people browsing random websites.” It is heavy flows to a relatively small number of hyperscale data centers and cloud regions. Those data centers are on land, connected by thick terrestrial fiber. Even a satellite network still needs dense ground gateways and terrestrial backhaul to land traffic where compute and storage live.

This is why “all internet moves to space” runs into a basic issue. Space networks still depend on Earth networks for the parts that carry the most bits.

Latency flips from a weakness to a niche advantage

Traditional geostationary satellite internet has high latency, often around the 500 to 600 millisecond range, because the signal travels so far.

LEO changes that, that is why Starlink moved their satellites lower. With satellites much closer to Earth, measured median latencies in Europe have been reported around the 40 millisecond range in some countries.

That is good enough for video calls and many cloud apps. In some specific long distance routes, LEO with optical inter-satellite links can even compete with fiber on latency because light travels faster in vacuum than in glass.

The catch is that those wins are route-specific and demand a mature mesh of laser links and ground stations.

So latency does not force the internet into space. That is why World Liberty Financial wants to move DeFi into space. Finance, defense, aviation, maritime logistics, and edge compute are all latency sensitive in ways that justify paying more for “good enough everywhere” or “fast on this corridor.”

The next phase is a hybrid internet, and it reshapes power

What changes the strategic picture is the emergence of space as a parallel distribution layer that can bypass chokepoints and shift bargaining power.

Two signals matter.

One is scale and regulatory momentum. The FCC just approved SpaceX to deploy an additional 7,500 second-generation Starlink satellites, with milestones into 2028 and 2031, while noting the system is still evolving. That is the shape of an infrastructure program, not a niche ISP.

The other is competition moving upmarket. Blue Origin’s newly announced TeraWave is explicitly pitched at enterprise, government, and data center class customers, not just rural households.

That is a bet that space connectivity becomes part of critical infrastructure and cloud adjacency, not merely consumer broadband. Find out here how much of a threat is TerraWave to Starlink.

Meanwhile, the physical risk surface also changes. Subsea cables are increasingly discussed as geopolitical and resilience assets because so much global traffic depends on them.

The reality of the future of the internet

Most internet traffic will stay on Earth because fiber is too efficient at moving bulk data, and the cloud lives on land.

But space will become a standard way to reach the internet and to keep it running when terrestrial links are slow, scarce, attacked, or simply not worth building.

The winning architecture is not “internet in space.” It is a hybrid internet where space is always on. That means networks can route around land-based operators and governments.