How 600 Undersea Cables Route the Global Internet Through Three Physical Chokepoints

At a depth of 3,000 meters in the Luzon Strait, where the Philippine Sea meets the South China Sea, a bundle of fiber-optic cables no thicker than a garden hose rests on the ocean floor. Inside these steel-armored tubes, light pulses carry the vast majority of digital communication between Japan, Taiwan, and Southeast Asia. From the surface, the strait appears as open water, but on the seabed, it is one of the most congested and critical infrastructure corridors on Earth.

It is a common misconception that global communication is accomplished primarily via satellite. In reality, over 99 percent of international data and voice transfers are routed through the physical fiber-optic cables that crisscross the world's seafloors. According to the Federal Communications Commission, the U.S. regulatory agency overseeing the infrastructure, these lines "provide the primary means of connectivity – voice, data and Internet – between the United States and the rest of the world," carrying a majority of civilian, military, and government offshore communications traffic. The total network spans over 1.5 million kilometers globally, connecting every continent except Antarctica.[2][3]

The physical construction of a modern submarine cable is remarkably delicate given the environment it inhabits. The core consists of glass optical fibers, roughly the diameter of a human hair, through which lasers fire data at extremely rapid rates using Dense Wavelength Division Multiplexing. These fibers are encased in a copper or aluminum tube that carries high-voltage electricity to power the repeaters spaced along the route. This core is then wrapped in layers of polycarbonate, water barriers, stranded steel wires for armor, and an outer sheath of polyethylene.[2]

While the transatlantic route between North America and Europe possesses massive redundancy—with dozens of cables spread across a wide geographic expanse—other regions are structurally constrained by geography. The connection between Europe and Asia, for example, is almost entirely dependent on the Red Sea. Fifteen submarine cables pass through the narrow Bab el-Mandeb Strait, the southern mouth of the Red Sea that separates East Africa from the Arabian Peninsula. This creates a single point of failure for the data traffic of multiple continents.[4]

The vulnerability of these chokepoints is not theoretical. In early 2024, a vessel struck by a Houthi missile in the Red Sea dropped its anchor and drifted for two weeks before sinking. The dragging anchor severed three major cables, severely impacting internet connectivity across East Africa and South Asia. Repairing the damage took months, as cable repair ships require specialized equipment, secure conditions, and permits from multiple jurisdictions to operate in contested waters.[4]

The Luzon Strait faces a different set of physical threats. Located in the Circum-Pacific Ring of Fire, the seabed is prone to intense seismic activity. In December 2006, the Hengchun earthquakes off the southern tip of Taiwan severed six of the seven cable systems in the corridor. International communication and digital activity in Taiwan and neighboring nations was instantly interrupted; banks could not process transactions, and airlines were forced to halt check-ins. The event demonstrated how quickly the cloud collapses when its physical foundation is fractured.[4]

The legal framework governing this infrastructure was drafted long before the internet existed. The United States and other nations rely on the United Nations Convention on the Law of the Sea (UNCLOS), specifically Article 112, which guarantees the right of all states to lay submarine cables and pipelines on the bed of the high seas beyond the continental shelf. However, UNCLOS provides little enforcement mechanism against accidental damage, which accounts for the vast majority of cable faults.[1]

Fishing trawlers and ship anchors cause over 70 percent of all cable faults globally. When a cable is severed, the repair process is entirely physical. A specialized cable-laying ship must navigate to the fault location, deploy a remotely operated vehicle or a grappling hook to retrieve the broken ends from the seabed, bring them to the surface, splice the microscopic glass fibers back together on the ship's deck, and lower the repaired section back into the ocean.[2]

The economics of the network are currently undergoing a structural shift. Historically, submarine cables were built by consortiums of state-owned telecommunications companies. Today, hyperscale cloud providers—specifically Meta, Google, Microsoft, and Amazon—are the dominant investors. Driven by the massive bandwidth requirements of their data centers, these four companies now own or lease the majority of new transoceanic capacity, shifting control of the internet's physical layer from national carriers to private technology firms.[4]

To mitigate the risks of the Red Sea and Luzon Strait chokepoints, these companies are attempting to diversify routes. New projects aim to bypass the South China Sea entirely by routing cables through the Celebes and Banda Seas, connecting Singapore directly to the United States via Guam. Other initiatives seek to connect Europe and Asia via terrestrial routes across the Arabian Peninsula or through the Arctic Ocean, though these face their own geopolitical and environmental hurdles.[4]

Despite these efforts, the fundamental reality of the global internet remains unchanged. It is not a decentralized, ethereal cloud, but a highly centralized, physical network of glass and steel resting on the ocean floor. As long as geography dictates the most efficient paths for these cables, the global economy will remain tethered to the stability of a few narrow, vulnerable straits.[4]