Abstract
Undersea communication cables have historically operated within an interconnected global system characterized by cooperative redundancy and routing agreements, despite episodic geopolitical contestation. Intensifying great power rivalry, however, is driving efforts to reroute cables, exclude vendors and build “trusted corridors.” Drawing on recent incidents and policy initiatives, this paper argues that emerging bifurcation threatens to weaken resilience by reducing redundancy, narrowing routing options and increasing incentives for disruption. While verified sabotage remains limited, securitization discourse is reshaping network design in ways that may prove counterproductive. The paper concludes with pragmatic policy options — including minilateral coordination and a scalable regional seas approach — to strengthen cable security while preserving interconnection.
Introduction
Undersea communication cables are the backbone of the global digital economy. Laid across the seabed in dense, intersecting routes, they carry more than 95 percent of all intercontinental data traffic, enabling everything from financial transactions and cloud services to online meetings and secure communications. As global data demand skyrockets, so too does our reliance on this largely invisible infrastructure.
Undersea cables are fibre-optic cables sheathed in layers of steel wire armouring, copper conductors and a waterproof polyethylene jacket, and about the circumference of a garden hose. They transmit light signals through bundled glass strands at nearly the speed of light. From one end of an ocean to another, they terminate at cable landing stations — onshore facilities located within a state’s sovereign jurisdiction. Here, the cables connect to terrestrial fibre, data centres and the broader national communications network.i
For decades, the undersea cable ecosystem has proven remarkably resilient, particularly in peacetime and outside moments of major interstate war. Redundant cable connections, reciprocal routing, cost-sharing for repairs, and public–private operating practices have ensured security and reliability.1
What follows explains how the increased sabotage and securitization of cables have resulted in fragmentation of the cable ecosystem — and why some responses risk undermining the very resilience that has served the system well.
From Obscurity to Securitization
Interdependence is now at risk for two reasons.
First, instances of cable sabotage, albeit infrequent, are increasing. In the Red Sea, at least four cables were cut in early 2024, amid escalating attacks on international shipping and maritime infrastructure by the Houthi in Yemen. Although the cable cuts were officially attributed to anchor drag, the circumstances and location of the cuts raised suspicions about intentional breakage, especially after two more cables were severed in September 2025 in the Red Sea near Jeddah, Saudi Arabia.2
In the Pacific near Taiwan, in early 2023, repeated cable disruptions near the Matsu Islands were initially framed by Taiwan as possible deliberate interference involving Chinese vessels,3 although no formal sabotage determination was made. More recently, incidents near Penghu — Taiwan’s main offshore archipelago — have reinforced concerns about grey-zone pressure on digital infrastructure.4 While these cases have not resulted in sabotage convictions under international law, they have nevertheless shaped threat perceptions, policy planning and defensive postures. On the other side of Eurasia, two separate cable systems were damaged in the Baltic Sea in late 2023 — one linking Finland to Estonia, the other connecting Sweden and Estonia.5 The following year, in November, the C-Lion1 cable connecting Finland and Germany was severed.6 Almost at the same time, and also in the Baltic, the BCS East–West Interlink cable running between Lithuania and Sweden was also cut.7 This affected only these two states, with Lithuania the more affected because the cable supplies one-third of its internet bandwidth. Several governments and security analysts have publicly framed these incidents as sabotage linked to hybrid operations.8
Second, as great power rivalry intensifies, a growing number of states and technology companies are rethinking how and where cables should be laid and how best to secure them against sabotage and espionage.9 This strategic decoupling — part of wider efforts by Canada, the European Union, Japan, the United States and others to de-risk from China — signals a shift toward bifurcation that leads away from the existing interconnected, global undersea cable ecosystem. This shift is engendered by the growing distrust between rivals in the international system and the quest by certain states (Australia, Japan, the United Kingdom and the United States) and groups of states (such as the European Union and the Quad — Australia, India, Japan and the United States) to improve the security of critical infrastructure.
The motivation to bifurcate and fragment undersea cable interconnectivity and redundancy, however, rests on flawed assumptions. Fragmentation will not only create new vulnerabilities but also undermine the resilience that has made cables largely secure in the first place.
Cables as Security Infrastructure
For decades, undersea cables operated in relative obscurity, treated as technical infrastructure managed by a mix of private consortia and lightly regulated by national or international regimes such as the 1884 Convention for the Protection of Submarine Telegraph Cables and the 1982 UN Convention on the Law of the Seas. That is no longer the case. As the geopolitical temperature rises, undersea cables are being securitized by state actors,ii recast not just as data pipelines but as critical and strategic infrastructure whose control, protection or disruption can do serious damage to the national security of a state or group of states.10
In recent years, governments have begun integrating cable protection into broader frameworks of economic and national security.11 In the United States, for instance, cable vulnerabilities are now discussed in the context of strategic competition with China alongside semiconductor chips, deep-sea mining and rare earth minerals.12 Accordingly, in 2020, the United States launched the Clean Network initiative, which asked countries and companies to abide by a set of shared principles in technology adoption, data privacy and security practices.13 It also built on previous efforts by Washington to encourage states to choose cable operators and builders that were non-Chinese.
In practical terms, this has steered new consortia away from Chinese vendors and jurisdictions perceived as high risk. In September 2025, the US House of Representatives passed legislation to tighten US control over critical fibre optic undersea cable equipment, in an attempt to prevent rivals such as China from acquiring technologies used in undersea cables.14
Parallel developments have occurred within multilateral and minilateral frameworks. In May 2023, the Quad Partnership for Cable Connectivity was announced. This was meant to underscore the strategic importance of secure, diversified digital infrastructure across the Indo-Pacific. Framed as a regional development priority, the initiative also reflected a growing perception of cables as potential targets of coercion and disruption and further embedded them within the broader context of strategic competition, rivalry and shifting distributions of global power.15
Taking the lead in the Quad’s effort, Australia stood up a Cable Connectivity and Resilience Centre to share best practices and provide technical assistance across the Indo-Pacific.16 Japan, meanwhile, has elevated cables within its economic security agenda, backing budget measures to diversify routes and landing stations, as well as, more recently, considering subsidies for cable-laying ships to secure national capacity in deployment and repair.17
In Europe, NATO has launched the Critical Undersea Infrastructure Network and established the Maritime Centre for Security of Critical Undersea Infrastructure to coordinate efforts across allies. In the Baltic region, a memorandum of understanding signed by NATO members and the European Union commits to enhanced cooperation in defending undersea infrastructure.18
By elevating cables within national and minilateral security agendas, states have increased the political salience of the network — raising the stakes of disruption even as protection efforts expand. But these efforts to secure cables have also extended into elite policy discourse and academic commentary that explicitly securitizes cables as objects under threat that need securing. Policy and scholarly debates19 increasingly invoke Henry Farrell and Abraham Newman’s “weaponized interdependence” framework20 to interpret cable exposure — especially the risks of monitoring (panopticon) and denial (chokepoint) associated with dependence on rival-linked infrastructure. Academics, while offering more nuance, nevertheless draw on the logic of weaponized interdependence to argue that states should decouple from infrastructure tied to strategic competitors.21 A similar logic of avoiding exposure and reducing risk — no matter the cost — has begun to surface in policy circles pushing bifurcation of cable routes. In practice, this has encouraged “trusted route” planning that skirts contested waters and rival jurisdictions.
Accordingly, governments and companies have begun proposing and funding new cable routes that avoid landfall on the sovereign territories of strategic competitors. The Echo and Apricot cable systems, jointly backed by Google and Meta, are designed to avoid landfall in China and routing through parts of the South China Sea.22 More broadly, big tech firms such as Google, Meta, Microsoft and Amazon have become the dominant financiers of new cable deployments, driven by escalating demand for data capacity, control and security.23 They have also been pressured by the US government to avoid using Chinese-made technologies in undersea cables.24 This matters because the infrastructure at stake is predominantly privately owned and operated; as Lars Gjesvik shows,25 contemporary network coercion often works through private corporate infrastructure and the regulation of firms, rather than through direct state ownership of the underlying networks.
Securitization and the Bifurcation Trap
This turn to route avoidance and trusted corridors is often justified through the logic of weaponized interdependence: policymakers increasingly presume that undersea cables will generate the familiar effects associated with network power — namely, chokepoint leverage, heightened exposure to coercion or denial, and strategic vulnerability through dependence on rival-linked infrastructure. These concerns are not unfounded: espionage risks associated with telecommunications infrastructure are well documented, and competition over cable supply chains reflects broader efforts by states to manage dependency, influence standards and limit exposure to rival-linked firms. For submarine cables, however, those effects are frequently overstated in peacetime conditions, because dense redundancy, cooperative routing arrangements and the boomerang costs of disruption make the system difficult to weaponize without wide collateral impact. It is this extension of network power logic — rather than the operational characteristics of cables themselves — that has encouraged bifurcation.
The cumulative effect of these efforts and the securitization of cables is paradoxical. While intended to insulate networks from coercion, the building of corridors that avoid connections and landfall with certain countries risks hardening the system into rival blocs. Bifurcation is likely to undermine the system’s core resilience, which has long derived from interdependence, such as route diversity, reciprocal restoration and an interlinking architecture that cushions shocks and keeps data flowing. More specifically, bifurcation does not refer to a single process but to a layered reconfiguration of the undersea communication cable ecosystem. It encompasses the fragmentation of cable routes to avoid specific jurisdictions, the exclusion of certain vendors and suppliers, the segmentation of financing and ownership structures, and the emergence of parallel governance and security frameworks.
These shifts are costly. Building parallel routes, excluding suppliers and maintaining duplicated landing infrastructure raises capital and operational expenditures, lengthens deployment timelines, and reduces economies of scale. For private firms — which finance most new cable systems — these costs translate into higher prices, constrained redundancy, or delayed upgrades, particularly outside core markets.26 Cost pressures also matter for security outcomes. Expensive, politically constrained networks tend to be thinner, more brittle and slower to repair, especially in regions with limited traffic volumes. In this sense, bifurcation risks trading perceived geopolitical insulation for reduced operational resilience.
The temptation to “nationalize” data routes is not new. It harks back to the old imperial cable networks built by imperial powers over a century ago. Russia’s announcement of efforts to establish a sovereign internet — routing domestic traffic through state-controlled exchanges and limiting foreign linkages — illustrates the short-sightedness of such strategies. Rather than insulating the system from external risk, such strategies create smaller, bounded networks that transmit less data and are easier to tap, sabotage and isolate.27 The same holds true for undersea cables: a bounded cable system sacrifices redundancy in the name of security and thereby heightens vulnerability to sabotage because potential adversaries will no longer share any blowback effect. In short, fragmentation narrows options and magnifies shocks; interconnection distributes risk and dampens escalation incentives.
Available evidence and experience suggest that the current, interconnected system can, under peacetime conditions, reduce incentives for deliberate disruption, insofar as interference risks widespread collateral effects across multiple states and commercial actors. This can include, in some cases, the initiator.28 Severing one system can degrade bandwidth and increase latency not only for an intended rival, but for multiple third parties, including, in some cases, the saboteur itself. The resulting boomerang effect acts as a deterrent against deliberate sabotage, particularly in peacetime. Conversely, when networks are segmented into “ours” and “theirs,” the anticipated blowback diminishes, making disruption comparatively more attractive.
This does not imply that interdependence guarantees restraint, nor that actors will always avoid self-harm; rather, it indicates that under conditions of dense redundancy and shared routing, undersea cables are less readily weaponized as instruments of denial than is often assumed. Adding to this, empirical evidence of legally verified sabotage against undersea communication cables, to date, remains limited. Most cable cuts still result from routine causes — anchoring, fishing gear or natural events — rather than hostile acts.29 At the same time, several recent incidents — particularly in the Baltic Sea and around Taiwan — have been politically framed as potential acts of deliberate interference, even where definitive attribution has proven elusive. This ambiguity is not incidental: the depth, remoteness and technical complexity of cable systems make conclusive forensic determination exceptionally difficult. As a result, undersea cable incidents increasingly operate in a grey zone, where suspicion, securitization and strategic signalling matter as much as legal proof.
Nevertheless, in open, highly networked systems such as the North Sea, Mediterranean or the western Pacific, the impact of any single cable break is limited. While damage can cause temporary latency and service degradation, redundancy and routing agreements among operators generally ensure continuity of data flow. Importantly, data does not flow automatically through the network: rerouting during disruptions depends on pre-existing commercial agreements, capacity and interoperability among operators, rather than on physical connectivity alone. Recent Baltic disruptions, for instance, also prompted data rerouting with limited functional impact. This underscored cable operators’ and the ecosystem’s adaptive capacities without minimizing the risks of multiple, simultaneous incidents.30 The same interdependence that securitization discourse now frames as vulnerability has, in practice, functioned as resilience.31
Rather than pursuing fragmentation, a more effective response would preserve and enhance interconnection, geographic dispersion and cooperative governance. These factors remain the strongest bulwarks against sabotage, systemic failure and strategic miscalculation.
Avoiding Fragmentation; Maintaining Resilience
Dense, redundant networks enable rerouting and limit disruption. Connectivity — not bifurcation — underwrites resilience, while fragmentation risks brittleness. The policy prescriptions below outline pragmatic steps to strengthen cable security without sacrificing interconnection.
- National actor mapping and points of contact: Identify who does what in cable security — from regulators and navies to private operators and repair contractors — to ensure rapid coordination during outages.
- Region-specific repair and monitoring protocols: Catalogue the locations, capacities and mobilization times of cable repair ships and formalize response frameworks between operators and governments.
- Minilateral cooperation frameworks: Encourage small, like-minded groupings (e.g., the Quad) to harmonize standards, share intelligence and coordinate repair logistics.
- Regional seas approach as a scalable governance model: Maritime security is spatially variant and thus regionally anchored security strategies across multiple maritime scales — coastal, zone, basin — are optimal.32 If we segment the seas into zones, actors can coordinate in contextually meaningful maritime zones. This “regional seas” approach — treating maritime security as basin-specific, coordinating states and operators within contextually meaningful sea spaces to manage cable protection, monitoring and response in ways that complement global governance — requires creativity and could include accepted basins such as the Baltic Sea or the Sea of Japan.33 Alternatively, regional seas could be conceptualized around existing undersea cable routes — such as a Pacific loop linking Hawaii, Guam, Japan, Taiwan and the Philippines. This would offer a geographically bounded governance framework, but one that also fits into global governance frameworks.
Conclusion
Undersea cables have moved from invisibility to the centre of geopolitical discourse, encouraging trusted corridor policies that risk hardening the system into rival blocs. A safer path is to strengthen interdependence while improving protection through mapped points of contact, repair capacity, minilateral coordination and a scalable regional seas approach.
