In 2023, a series of undersea cable cuts in the Red Sea disrupted internet connectivity across East Africa and parts of South Asia for weeks. The incident barely registered in Western media cycles. Yet it exposed something fundamental: the global information environment rests on a surprisingly fragile physical substrate, one shaped by colonial trade routes, Cold War strategic calculations, and the capital allocation decisions of a handful of technology conglomerates.

We tend to think of media in terms of content—what gets made, who watches it, which narratives dominate. But beneath every streaming service, every newsroom's cloud-hosted CMS, and every social platform's recommendation engine lies a layer of physical and logistical infrastructure that determines whether content can move at all. Cables on the ocean floor. Data centers consuming the electrical output of small cities. Regulatory regimes that dictate where data can be stored and what content can be served. This is the infrastructure layer, and it is deeply, irreducibly geopolitical.

Harold Innis argued that the material characteristics of communication media—their durability, portability, and the resources required to produce them—shape the structure of the civilizations that depend on them. That insight has never been more relevant. Today's media infrastructure creates dependencies between nations, concentrates leverage in specific corporate and state actors, and produces vulnerabilities that carry genuine strategic consequence. Understanding who controls this infrastructure, and how, is no longer optional for anyone serious about media systems.

Approximately 95 percent of intercontinental data traffic travels through roughly 550 submarine fiber-optic cables laid across the world's ocean floors. These cables, most no thicker than a garden hose, carry the financial transactions, media streams, and communications that constitute the operational backbone of the global economy. Their routes are not neutral. They follow patterns established by telegraph cables in the 19th century, which themselves traced colonial shipping lanes. The result is an information topology that still privileges connections between former imperial centers and their historical peripheries.

Ownership of these cables has shifted dramatically over the past decade. Where telecommunications consortia once dominated submarine cable investment, technology companies—primarily Google, Meta, Microsoft, and Amazon—now own or lease substantial portions of newly laid transoceanic capacity. Google alone has invested in cables spanning the Atlantic, Pacific, and routes connecting South America and Africa. This represents a vertical integration of remarkable scope: the same companies that operate the dominant content platforms and cloud services now control significant portions of the physical layer those platforms depend on.

The strategic vulnerability this creates is not hypothetical. Cable cuts—whether from anchors, earthquakes, or deliberate sabotage—can isolate entire regions. Nations with limited cable landing points, often in Africa and the Pacific Islands, face acute dependency. When a cable serving Tonga was severed in 2022 following a volcanic eruption, the nation was effectively disconnected from the global internet for weeks. Military and intelligence agencies in multiple countries have identified submarine cables as critical infrastructure targets, and several nations have invested in cable surveillance capabilities.

Redundancy is unevenly distributed. Major hubs like Marseille, Singapore, and Mumbai benefit from multiple cable landings and diverse routing options. But the global cable map reveals chokepoints—the Suez Canal corridor, the Strait of Malacca, the waters around Cornwall—where concentrated cable routes create systemic risk. A sufficiently disruptive event at any of these points would degrade media distribution capacity for hundreds of millions of users simultaneously.

The geopolitical implications extend beyond vulnerability to active competition. China's investments in new cable routes through the Peace Cable and other projects represent an effort to build information pathways that bypass traditional Western-controlled chokepoints. The United States has intervened to block Chinese participation in cables connecting to its territory, citing national security concerns. What emerges is a picture of the ocean floor as genuinely contested territory—an arena where infrastructure investment decisions carry strategic weight comparable to military positioning.

Takeaway

The physical routes through which information travels are neither neutral nor inevitable. They encode historical power relationships and create dependencies that constrain media systems as powerfully as any editorial decision or content regulation.

Modern media production and distribution have migrated almost entirely to cloud infrastructure. Newsrooms draft, edit, and publish through cloud-hosted content management systems. Streaming platforms encode and deliver video from cloud data centers. Social media feeds are assembled by algorithms running on rented compute capacity. Three providers—Amazon Web Services, Microsoft Azure, and Google Cloud—control approximately two-thirds of the global cloud infrastructure market. This concentration creates a dependency relationship that reshapes the economics and politics of media in ways that are still poorly understood.

The implications begin with geography. Cloud providers locate their data centers based on energy costs, cooling efficiency, tax incentives, and proximity to network interconnection points. These decisions determine the latency and performance characteristics of media services in different regions. A streaming platform serving Southeast Asia from data centers concentrated in Northern Virginia operates under fundamentally different constraints than one with regional infrastructure. The result is an uneven distribution of media delivery quality that maps onto existing economic inequalities.

More consequentially, cloud concentration creates a platform dependency distinct from the content-layer platform power that receives most regulatory attention. When a cloud provider experiences an outage—as AWS did in a significant incident in December 2021—entire constellations of media services go dark simultaneously. Newspapers, streaming services, podcasting platforms, and e-commerce sites all failed together, revealing their shared dependence on a single infrastructure provider. This correlated failure mode has no precedent in the history of media distribution.

The economic dynamics reinforce concentration. Cloud providers achieve margins through massive scale, making it increasingly difficult for smaller competitors to offer comparable price-performance. Data gravity—the tendency for applications and services to cluster around existing data stores—creates powerful lock-in effects. Once a media company's archives, production pipelines, and distribution systems are built on a particular cloud platform, the cost of migration becomes prohibitive. This gives cloud providers structural leverage over the media companies that depend on them, a form of power exercised through pricing, terms of service, and infrastructure roadmap decisions rather than editorial intervention.

This is not solely a commercial matter. Governments have recognized that cloud infrastructure concentration carries national security implications. The U.S. Department of Defense's multi-billion-dollar cloud contracts, and the European Union's Gaia-X initiative to develop sovereign cloud capacity, reflect an understanding that dependence on a small number of cloud providers—predominantly American—creates strategic vulnerability. For media systems specifically, the question is whether the infrastructure layer can be governed in ways that preserve editorial independence and distribution diversity when the underlying compute and storage is controlled by entities with their own commercial and political interests.

Takeaway

When a handful of cloud providers serve as the operational foundation for most of the world's media, infrastructure decisions that appear purely technical—where to build data centers, how to price storage, which services to prioritize—become de facto media policy.

The vision of a single, borderless internet as the distribution medium for global media has been quietly replaced by a reality of increasing regulatory fragmentation. Data localization laws—requirements that certain categories of data be stored and processed within national borders—have been enacted or proposed in over 70 countries. Content regulation regimes vary dramatically, from the European Union's Digital Services Act to India's IT Rules to China's comprehensive content control apparatus. For media companies operating at scale, this fragmentation is now the defining structural challenge.

Data localization creates direct infrastructure costs. When a country requires that user data or content assets be stored domestically, media companies must either build or lease local data center capacity, negotiate with approved local cloud providers, or exit the market entirely. India's data localization provisions, for instance, have forced global streaming platforms to invest in domestic infrastructure and negotiate complex compliance arrangements. These requirements effectively create national media infrastructure zones, each with its own cost structure and technical constraints.

Content regulation adds a parallel layer of complexity. The EU's audiovisual media services directive requires streaming platforms to ensure that at least 30 percent of their catalog consists of European works. Several countries mandate local language content quotas or require editorial review processes for user-generated content. These regulations do not merely affect what content is available—they reshape the economics of content production by creating guaranteed demand for regionally specific material. The infrastructure to manage compliance—geo-fencing content, implementing region-specific recommendation algorithms, maintaining multiple content catalogs—becomes itself a significant operational and architectural challenge.

The strategic effect of this fragmentation is a gradual re-nationalization of media distribution infrastructure. Companies with the resources to navigate regulatory complexity—typically the largest global platforms—gain competitive advantages over smaller or regional players who cannot absorb compliance costs. Paradoxically, regulations intended to promote local media diversity can reinforce the dominance of well-resourced global incumbents who can afford the infrastructure investment that compliance demands.

What emerges is a media distribution landscape shaped as much by regulatory topography as by audience demand or content quality. The splinternet thesis—that the global internet is fragmenting into distinct regulatory zones—is no longer speculative. It is the operating reality for any media company with international distribution. The geopolitical dimension is explicit: regulatory frameworks reflect national strategic priorities, and the fragmentation of media infrastructure along national lines is both a consequence and an instrument of great power competition over information flows.

Takeaway

Regulatory fragmentation does not simply constrain media distribution—it actively restructures it, creating national infrastructure zones that reshape what content gets made, where it can travel, and who has the resources to navigate the resulting complexity.

Media infrastructure is not a backdrop to the more visible drama of content creation and audience engagement. It is the primary terrain on which the future of global information flows is being determined. Submarine cables, cloud platforms, and regulatory regimes constitute a layer of power that operates beneath the threshold of most public discourse about media—yet shapes it more decisively than any algorithm or editorial choice.

For media professionals and policymakers, the strategic imperative is clear: develop infrastructure literacy. Understanding who owns the cables, who operates the clouds, and how regulatory fragmentation reshapes distribution economics is now essential to any serious analysis of media power. The decisions being made at this layer will determine which societies have resilient, diverse information environments and which are dependent on infrastructure controlled by distant actors.

The geopolitics of media infrastructure is not a future concern. It is the present condition of every media system on the planet. The question is whether it will be governed deliberately or allowed to consolidate by default.