Nearly all of the world’s internet traffic travels through cables on the ocean floor

Satellites tend to get the credit for connecting the world, but the internet’s real backbone sits at the bottom of the ocean. A sprawling network of fiber-optic cables, some no thicker than a garden hose once past the shoreline, carries almost every email, video call and financial transaction that crosses an ocean. Satellites handle only a small sliver of that traffic, despite the popular image of data beaming down from space.

Thin Cables Carrying Nearly the Whole Internet

A submarine communications cable is a bundle of glass fibers, protected by layers of insulation and armor, laid directly on or buried beneath the seabed to carry data between continents. More than 600 of these cables currently crisscross the world’s oceans, stretching a combined distance of roughly 1.48 million kilometers, or about 920,000 miles, enough to circle the planet dozens of times over. Despite decades of advances in satellite technology, industry analysis from telecommunications research firm TeleGeography puts the share of intercontinental data and voice traffic carried by these undersea lines at somewhere between 95 and 99 percent, making satellites a supplementary system rather than a genuine alternative for bulk data. Satellites remain valuable for reaching ships at sea, remote research stations and areas with no cable landing nearby, but the physics of radio transmission over thousands of miles of vacuum and atmosphere introduces delays and bandwidth limits that fiber simply does not face, which is why even the newest low-orbit satellite constellations have not displaced cable as the backbone of global data traffic.

From Telegraph Wires to Fiber-Optic Backbones

Undersea cables are not a recent invention. The first successful transatlantic telegraph cable went into service in the mid-19th century, carrying slow electrical pulses rather than digital data, and cable technology advanced steadily through coaxial copper systems over the following century. The shift that created the modern internet backbone came with fiber optics, which replaced electrical signals with pulses of light traveling through hair-thin glass strands capable of carrying vastly more information with far less signal loss over long distances. That fiber-optic transition let a single cable carry the equivalent of what once required dozens of older copper lines, setting up the dense global mesh that now exists. The first fiber-optic transatlantic cable entered service in the late 1980s, and each generation of cable technology since has packed in more wavelengths of light traveling simultaneously down the same glass strands, multiplying capacity many times over without needing to lay an entirely new physical line for every increase in demand.

Who Owns the Modern Cable Network

Ownership of this infrastructure has shifted dramatically in the past decade. Undersea cables were historically built and operated by telecommunications carriers working in consortiums, but major technology companies now fund and control a growing share of the network directly. According to the submarine cable industry’s own tracking, Google, Meta, Amazon and Microsoft together now control more than 70 percent of transoceanic cable capacity, a reflection of how much of global data traffic is generated by their own cloud computing and streaming services rather than by traditional phone and internet providers. Where a single telecom consortium once might have spent years assembling a syndicate of national carriers to fund a new route, a cloud provider can now commission and own an entire cable outright, tailoring the route and capacity specifically to connect its own data centers across an ocean rather than serving the general carrier market first.

The Fragility Beneath the Traffic

For infrastructure this critical, the physical cables themselves are surprisingly vulnerable. Sections running through shallow coastal waters are often no thicker than a garden hose, and the most common cause of damage is not sabotage or natural disaster but fishing trawlers and ship anchors dragging across the seabed. Earthquakes and underwater landslides have also severed major cables in the past, in some cases cutting off large portions of a region’s international connectivity for days at a stretch. A notable disruption in 2008 damaged several Mediterranean cables in quick succession, slowing internet access across parts of the Middle East and South Asia and demonstrating how concentrated some of these choke points remain even in a network with hundreds of individual lines. Cable faults of one kind or another are actually a routine occurrence rather than a rare event, with dozens of breaks reported and repaired worldwide in a typical year, most of them minor and invisible to ordinary internet users because traffic automatically reroutes through unaffected lines nearby.

Keeping the Network Repaired and Running

Because outages carry such an outsized economic cost, the industry maintains a standing fleet of specialized cable repair ships stationed around the world, ready to sail out, locate a break using onboard sensing equipment, haul the damaged section to the surface and splice in a new length of fiber. Repairs can take days to weeks depending on weather, water depth and how far the ship has to travel, which is part of why cable operators route new lines along diverse paths rather than clustering them in a single narrow corridor. That redundancy, built up gradually over decades of cable-laying, is a major reason a handful of localized breaks rarely bring down the wider internet, even though the vast majority of the traffic keeps moving through infrastructure that most users never think about and will never see. The specialized ships that perform these repairs are few in number relative to the size of the global network, and cable industry groups have periodically raised concerns about whether the current repair fleet could keep pace if damage ever struck multiple major routes in the same region at once, a scenario that has drawn growing attention from governments treating undersea cables as critical national infrastructure rather than purely commercial assets.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview