CSIS links Chinese

CSIS analyzed AIS tracks for 223 research, survey and related vessels from 48 countries between January 1, 2020, and December 9, 2025. [1]

Chinese-flagged vessels had a positive, statistically significant association with time spent near seabed infrastructure on the high seas, although the model explained only 1.3% of variation in the data. [1]

Two case studies showed repeated, survey-like movements near Pacific cable routes, but CSIS said those tracks do not prove sabotage or military intelligence collection. [1]

Cable-industry and international-law sources emphasize monitoring, repair capacity, route resilience and clearer enforcement rather than treating every vessel near a cable as hostile. [2] [4] [5]

Chinese-flagged research vessels spent a statistically larger of their observed time near subsea infrastructure than research vessels from 47 other countries, according to a Center for Strategic and International Studies analysis published October 8, 2026. The study examined 223 vessels from 48 countries using automatic identification system data tracked between January 1, 2020, and December 9, 2025. [1]

The finding describes a pattern in ship movements, not a confirmed campaign to damage or tap cables. CSIS said repeated proximity to telecommunications routes and survey-like tracks could support future military planning or gray-zone operations, while acknowledging that AIS data shows where a ship traveled rather than what it did underwater. [1]

A comparison of tracked time, not a ranking of total hours

The CSIS Futures Lab combined AIS tracks from General Atomics’ Optix platform with maps of exclusive economic zones, high-seas areas, gas pipelines, fiber-optic cables and mineral-prospecting regions. Researchers treated a vessel as being near infrastructure when its track came within one kilometer of a mapped cable or pipeline. They then estimated the of each vessel’s tracked time spent near that infrastructure on the high seas. [1]

The sample was deliberately constructed rather than a census of every research ship. It included fisheries research vessels, oceanographic research and survey ships, dive vessels, special vessels and fishery patrol vessels. Researchers verified vessel types and maritime mobile service identity numbers using MarineTraffic before analyzing the tracks. [1]

The comparison concerns the of observed time, not absolute time spent near cables. Most vessels spent less than 1% of their annual tracked time near mapped seabed infrastructure, according to a secondary review of the CSIS methodology. In 2024, seven Chinese research vessels exceeded the study’s 80th-percentile threshold. U.S. vessels also ranked near the top in some years, so the pattern is not unique to Chinese ships. [1] [2]

Chinese-flagged vessels remained statistically significant in the study’s robustness checks. In the baseline model, the China-flag variable had an estimated coefficient of 0.0315 with p<0.001; under more conservative specifications, the association weakened but remained near conventional significance thresholds. CSIS also reported that Chinese vessels appeared in the top 20th percentile for high-seas infrastructure activity in four of the five years studied, with a notable increase in 2024. [1]

The limitations are substantial. The model’s adjusted R-squared was 0.013, meaning the variables explained only a small portion of the variation in vessel behavior. The study also measured proximity to gas pipelines as well as telecommunications cables, and some cable routes are approximate. A cable crossing, repeated visit or AIS gap can have an ordinary explanation. [1]

Two Pacific case studies

The report focuses on the Xiang Yang Hong 06 and Ke Xue. Between February and May 2024, Xiang Yang Hong 06 made three trips to the high seas off Guam. Its parallel, repeated passes resembled a “lawnmower” pattern commonly associated with bathymetric surveying and occurred near mapped routes for the Apricot and AAG telecommunications systems. CSIS said the initial survey areas did not align closely with known mineral-prospecting areas, making cable infrastructure a relevant alternative explanation, while stopping short of calling the activity hostile. [1]

Ke Xue repeatedly operated west of Okinotorishima between March and May 2024, near or across the mapped routes of TGN-IA, Apricot and JUPITER. It later transited through the Sea of Japan and Tsugaru Strait toward the Bering Sea, where its tracks came near additional international cable systems, including Russia’s Polar Express. Some of the cables involved have Chinese ownership or participation, so the tracks do not show that the ship selected only foreign-controlled infrastructure. [1] [2]

The strategic concern is the information that survey work can produce. Bathymetry, sediment, temperature, salinity and current data have legitimate scientific uses but can also help submarines navigate, improve sonar performance and characterize the seabed. A separate U.S.-China Economic and Security Review Commission report describes trans-Pacific cables as critical infrastructure vulnerable to discreet, difficult-to-attribute disruption, while noting that most cable breakages are accidental or caused by natural hazards. [3]

No cable damage, tapping or other physical interference is established in the CSIS paper’s two case studies. The evidence is behavioral: location, duration, repetition and track geometry. That distinction matters because the report’s central statistical result and its individual-vessel narratives support monitoring and investigation, not a conclusion operational intent. [1]

The legal question depends on location

The legal picture changes sharply between the high seas and a coastal state’s exclusive economic zone. CSIS notes that surveying and data collection on the high seas are generally lawful under the United Nations Convention on the Law of the Sea. Within another country’s EEZ, marine scientific research requires coastal-state consent, but UNCLOS does not clearly resolve every distinction between marine science, hydrographic surveying and military survey activity. The United States and China take different positions on whether foreign military or hydrographic surveys require consent in an EEZ. [1]

That ambiguity limits the ability of coastal states to treat suspicious movement near a cable as an immediately punishable offense, particularly when a ship remains outside territorial waters and does not physically interfere with infrastructure. A legal review in the Asian Journal of International Law identified weaknesses in the regime governing cable damage and competing uses of the ocean, including enforcement and jurisdictional gaps. [2]

The high-seas setting is central to the CSIS analysis. The report’s statistical measure specifically isolates time spent near infrastructure outside EEZs, where coastal-state consent requirements are less directly applicable. That makes the activity harder to regulate, but not automatically unlawful. [1]

Resilience is broader than patrols

The cable industry’s response is broader than continuous patrols. A 2026 industry review drawing on the International Cable Protection Committee and regional cable organizations describes a layered approach involving burial, armoring, route engineering, maritime awareness, monitoring, repair capacity and cooperation between operators and governments. It also cautions that more than 80% of cable faults arise from routine maritime activity such as fishing and anchor drags, underscoring the need to distinguish suspicious behavior from the much larger background of accidental damage. [4]

The International Telecommunication Union and ICPC’s 2026 resilience work calls for better information-sharing between governments and operators, stronger AIS enforcement, improved fault investigations and clearer liability standards. Those measures address the evidence problem exposed by the CSIS study: authorities may detect a recurring pattern without being able to prove the mission behind a single voyage. [5]

CSIS recommends an international mechanism for sharing maritime-domain awareness, coordinated patrols near critical infrastructure and regional cable-repair hubs. The study supports closer observation of Chinese research fleets, but its own caveats argue against treating every survey pattern as proof of an impending attack. The confirmed development is a persistent monitoring signal around cables that connect U.S. territories and Asian partners across the Pacific. [1]