Concerns Over Chinese Undersea Cable Sabotage Grow as Trump-Xi Summit Approaches

On the ocean floor, buried in sediment or resting in trenches more than three kilometers deep, roughly 600 active fiber-optic cables carry 95% of the world's intercontinental data traffic [1]. They also move more than $10 trillion in financial transactions every day — a figure that TeleGeography, the industry's leading research firm, has verified by aggregating SWIFT messaging volumes with other payment networks [2]. As President Donald Trump arrives in Beijing on May 14 for a summit with Chinese President Xi Jinping, a growing chorus of defense officials, intelligence analysts, and legislators argue that this infrastructure has become a front line in great-power competition [3].

The question is whether the alarm is proportionate to the risk.

The Chokepoints Where Cables Meet Geopolitics

Of the roughly 600 submarine cables in operation, a disproportionate number pass through a handful of geographic bottlenecks. The Luzon Strait — the 250-kilometer channel between Taiwan and the Philippines — funnels at least 18 international cables connecting Northeast Asia to Southeast Asia and onward to the Americas [4]. The South China Sea, where China has built and militarized artificial islands, is traversed by cables linking Singapore, Vietnam, Hong Kong, and mainland China to the broader global network [5]. Egypt's Suez corridor, through which an estimated one-third of global internet traffic passes, constitutes another critical node [6].

In each of these chokepoints, the cables' physical proximity to military assets — Chinese naval vessels, coast guard patrols, and maritime militia fishing fleets — creates what security analysts describe as a permanent proximity threat. The Pacific Forum has characterized the South China Sea specifically as a zone where China's "cable-cutting threat" intersects with its broader territorial claims [5].

A Pattern of Incidents — and Contested Attribution

The record of cable disruptions near Chinese-affiliated vessels has grown steadily since 2020.

Source: Global Taiwan Institute / CSIS / media reports

Data as of May 11, 2026CSV

The most consequential incident to date occurred in February 2023, when two undersea cables serving Taiwan's Matsu Islands were severed, leaving 13,000 residents without internet access for nearly two months [7]. Taiwanese authorities traced the damage to a Chinese fishing vessel and a Chinese-flagged cargo ship, but stopped short of declaring the cuts deliberate [8]. Beijing called the incidents coincidental and accused Taipei of politicizing routine maritime accidents [7].

In February 2025, the Chinese-crewed vessel Hong Tai 58 damaged the TPKM-3 cable connecting Taiwan's Penghu Islands to the main island, prompting Taiwan's coast guard to detain the ship — a rare enforcement action [9]. Taiwan has reported approximately 30 subsea cable incidents in recent years, though not all involve Chinese-flagged vessels [10].

The pattern extends beyond the Indo-Pacific. In November and December 2024, multiple cables and a pipeline were severed in the Baltic Sea in incidents initially attributed to Russian hybrid warfare. The Chinese-flagged cargo vessel Yi Peng 3 was placed under scrutiny by European investigators, with Western intelligence officials reporting that the ship's anchor may have dragged across seabed infrastructure — though whether this was accidental or directed remained unresolved [11]. NATO launched its Baltic Sentry mission in response [12].

Yet the attribution picture is murkier than headlines suggest. In January 2025, the Washington Post reported that U.S. and European intelligence services had found no evidence that the Baltic cable damage was intentional, concluding that the incidents were likely maritime accidents [13]. Finnish lead investigator Sami Liimatainen publicly disputed the Post's characterization, noting he had not been contacted for the story [14]. Sweden subsequently closed one investigation, ruling it accidental, while other probes remained open [15].

This ambiguity is the core problem. Between 150 and 200 subsea cable faults occur every year worldwide, overwhelmingly caused by fishing trawlers and ship anchors [16]. Distinguishing a deliberate anchor drag from a negligent one requires evidence of intent that is extraordinarily difficult to establish at sea.

The $10 Trillion Cascade Scenario

The financial exposure figures circulating in policy circles trace back to a straightforward calculation. SWIFT alone processed roughly 45 million messages per day in recent years; combined with other clearing networks, the daily total comfortably exceeds $10 trillion [2]. These transactions depend on submarine cable connectivity because terrestrial and satellite alternatives lack the bandwidth and latency performance required for high-frequency trading, real-time clearing, and cross-border settlement.

The Financial Services Information Sharing and Analysis Center (FS-ISAC) released guidance in late 2024 warning member institutions that a coordinated multi-cable disruption could interrupt clearing systems, payment rails, and data routes that "lack terrestrial or satellite redundancy sufficient to sustain operations beyond 24 to 72 hours" in certain corridors [17]. FS-ISAC held a tabletop exercise on the scenario in March 2026, simulating the cascading effects on member firms [17].

The $10 trillion figure, however, represents daily transaction volume, not losses. The actual economic damage from a cable disruption depends on which cables are cut, for how long, and whether traffic can be rerouted. A single cable break is typically managed through redundant paths. A simultaneous, coordinated severing of multiple cables at a chokepoint like the Luzon Strait — which has no comparable terrestrial bypass — would be a different matter. CSIS has modeled such scenarios, finding that concentrated cable corridors strain the world's limited repair fleet if multiple faults occur simultaneously [18].

China's Own Vulnerability — and the Question of Deterrence

Any analysis of Chinese cable-cutting threats must reckon with a basic asymmetry: China is itself deeply dependent on the same infrastructure. Mainland China's international internet connectivity runs almost entirely through submarine cables, with landing stations concentrated along its eastern seaboard [19]. Chinese financial institutions, export logistics platforms, and Belt and Road communication networks all rely on cables that Beijing does not exclusively control [20].

HMN Technologies (formerly Huawei Marine), one of only four companies globally capable of manufacturing and laying transoceanic cables, gives China some industrial leverage. HMN can build cables at costs 20% to 30% below Western competitors [20]. But manufacturing cables and controlling the routes they traverse are different capabilities. The United States has blocked Chinese firms from subsea projects involving U.S. investment on national security grounds [20], and a Bloomberg investigation documented an intensifying U.S.-China competition over new cable routes, particularly those intended to serve AI data centers [21].

Does mutual dependence function as deterrence, analogous to nuclear mutually assured destruction? The analogy is imperfect. Unlike nuclear weapons, cable cuts are deniable, reversible (given enough time and repair ships), and graduated in effect. A state can sever one cable and observe the response before escalating. This makes cables more susceptible to gray-zone tactics — precisely the kind of ambiguous, below-threshold actions that China has employed in the South China Sea and around Taiwan [5].

The Repair Ship Deficit

The world's cable repair capacity is alarmingly thin. According to TeleGeography and Infra-Analytics, 62 vessels globally are engaged in cable laying and maintenance, but only 19 are contracted for dedicated repair service [22]. The fleet is aging: approximately two-thirds of cable maintenance ships will reach end of service life by 2040 [22]. The longest recorded cable repair took 947 days [23].

Source: TeleGeography / Infra-Analytics

Data as of May 11, 2026CSV

TeleGeography estimates that sustaining current service levels and avoiding repair delays will require roughly $3 billion in new investment, against approximately $800 million committed — a gap of $2.2 billion [22]. New cable ships take three to five years to commission, meaning that even if funding materialized today, the capacity shortfall would persist through at least the early 2030s.

The European Union Agency for Cybersecurity (ENISA) has warned that concentrated cable corridors can overwhelm repair capacity if multiple faults occur at once [23]. A coordinated attack on cables in the Luzon Strait, for example, could consume the entire regional repair fleet for months, leaving other damaged cables unattended.

Legislative and Diplomatic Responses

The U.S. legislative response has centered on the Strategic Subsea Cables Act of 2026, introduced by Senator Jeanne Shaheen and championed in the House by Representatives Joe Wilson and Gregory Meeks [24]. The bill would establish an interagency committee on undersea infrastructure, mandate at least ten dedicated State Department staff for subsea cable diplomacy, streamline permitting for new cable routes, and authorize sanctions against foreign persons who "knowingly sabotage or facilitate the sabotage of critical undersea infrastructure" [24].

Taiwan designated 10 domestic submarine cables as critical infrastructure in 2024, triggering heightened security and government oversight [10]. NATO established the Critical Undersea Infrastructure Network for threat intelligence sharing among public and private sector actors [12].

What remains conspicuously absent is any bilateral or multilateral agreement directly governing seabed infrastructure. Cold War-era precedents exist: the 1972 U.S.-Soviet Incidents at Sea Agreement established protocols to prevent naval confrontations from escalating, and similar frameworks governed behavior around undersea listening arrays. No comparable accord has been proposed for undersea cables, and the Trump-Xi summit agenda — centered on trade, AI, rare earth export controls, Taiwan, and the Iran conflict — does not appear to include seabed infrastructure specifically [3][25].

This gap matters. Without agreed norms, every cable cut triggers a cycle of accusation, denial, and ambiguity that makes escalation more likely.

The Steelman Case for Skepticism

Not everyone in the analytical community accepts the prevailing threat narrative at face value. A RAND Europe analysis, authored by a former submarine commander, argued that "attacking and disrupting the undersea cable network is not as easy as some reports might make it appear," characterizing deep-water sabotage as "difficult and risky" due to weather, currents, seabed conditions, and detection risk [16]. While China's Haiyang Dizhi 2 research vessel demonstrated a cable-cutting device at 3,500 meters depth in early 2026 — the first publicly acknowledged test of severing capability at that depth [26] — RAND notes that such operations require precise location data and specialist equipment that limit their scalability.

There is also a structural incentive question. Many of the loudest voices warning about cable sabotage are defense officials, intelligence community alumni, and think tanks whose funding overlaps with defense contractors positioned to benefit from infrastructure hardening contracts. The Strategic Subsea Cables Act itself would channel significant federal resources toward new infrastructure, permitting, and monitoring capabilities. This does not invalidate the threat, but it does mean that the assessors and the beneficiaries are sometimes the same people.

Skeptics further argue that cable sabotage fears serve a broader policy function: building public support for technological decoupling from China that would proceed regardless on trade and industrial policy grounds. The exclusion of HMN Technologies from U.S.-funded cable projects, for example, predates the current sabotage discourse and reflects longstanding concerns about Chinese access to communications infrastructure [20].

The Washington Post's January 2025 reporting on the Baltic incidents — in which multiple intelligence services concluded the cable damage was accidental — suggests that the inclination to attribute breaks to state sabotage may sometimes outpace the evidence [13]. At the same time, maritime experts who disputed that conclusion noted the implausibility of a ship accidentally dragging an anchor for more than 100 kilometers, illustrating how each incident generates more ambiguity, not less [14].

What the Evidence Supports

The factual picture, stripped of both alarmism and dismissiveness, looks roughly like this:

The infrastructure is genuinely vulnerable. Submarine cables were designed for commercial efficiency, not military resilience. They pass through identifiable chokepoints, land at known stations, and rely on a repair fleet that is too small and too old for current demand, let alone a crisis scenario [22][23].

China is developing purposeful capabilities. The progression from anchor-dragging incidents involving civilian vessels to a demonstrated deep-sea cutting device represents a qualitative shift that defense analysts have documented in detail [26][18].

Attribution remains the weakest link. No cable incident has been formally attributed to a state actor by the government of the affected country. The evidentiary threshold — proving intent behind an anchor drag, or linking a vessel's actions to state direction — exceeds what current maritime surveillance can reliably establish [11][13].

Mutual dependence is real but asymmetric. China needs international cables for its own economy, but its tolerance for disruption may differ from that of Western financial centers built on millisecond-latency connectivity [19].

The repair gap is the most concrete and least contested risk. Regardless of whether anyone deliberately cuts a cable, the mismatch between expanding cable infrastructure and shrinking repair capacity creates a systemic vulnerability that $2.2 billion in unfunded investment has not addressed [22].

As Trump and Xi meet in Beijing this week, the cables on the ocean floor beneath them will continue carrying the data and money that bind their economies together. Whether that shared dependence makes the cables safer — or makes them a more tempting target — is the question neither leader has yet been willing to answer publicly.