
In 2024, submarine disturbances near Ivory Coast simultaneously severed seven undersea cables, but for the Bitcoin network, this event left almost no trace: prices remained stable, consensus was unaffected. Cambridge researchers Wenbin Wu and Alexander Neumueller used this as a basis to analyze cable failure events from 2014 to 2025, concluding that “the real bottleneck is not underwater but in the cloud.”

(Source: Cambridge Research)
Cambridge’s dataset covers 11 years of Bitcoin network activity. Among 68 verified cable failure events, 87% caused node changes of less than 5%, with an average impact of -1.5%, and a median of only -0.4%. The correlation coefficient between node disruptions and Bitcoin prices is close to zero (r = -0.02), indicating almost no linear relationship.
Further simulation experiments quantify the actual threat of cable cuts: in scenarios where cables are randomly removed, over 72% to 92% of all cables must be cut to disconnect more than 10% of nodes—meaning an almost global-scale coordinated disruption of transoceanic fiber optic cables would be needed to significantly impact the Bitcoin network.

(Source: Cambridge Research)
However, when researchers shifted focus from physical cables to Autonomous System Numbers (ASNs), the figures changed dramatically. Precise attacks on high-centrality cable nodes can lower the critical threshold from 72-92% to just 20%; targeting the cloud service providers with the most nodes, removing only 5% of routing capacity can cause noticeable disruptions to reachable nodes on the public network.
According to a March 2026 Bitnodes snapshot, the hosting concentration of 23,150 reachable nodes is as follows:
Hetzner: 869 nodes (3.8%)
OVHcloud: 348 nodes (1.5%)
Comcast: 348 nodes (1.5%)
Amazon AWS: 336 nodes (1.5%)
Google Cloud: 313 nodes (1.4%)
The researchers explicitly state that this ASN scenario refers to “hosting providers’ shutdowns or coordinated regulatory actions, not physical cable cuts”—in other words, policy-level coordinated pressure rather than battlefield sabotage. Recent outages in Amazon and AWS Middle East regions, while not causing substantial impact on Bitcoin, confirm that such hosting failures are real risks, not just theoretical speculation.
Even with the above ASN risks, Bitcoin holds a widely underestimated trump card—Tor. Data from March 2026 shows that out of 23,150 reachable nodes, 14,602 (63%) operate over the Tor network, making it a major component of Bitcoin node infrastructure.
The surge in Tor usage was not planned but a collective response to a series of censorship events: Iran’s internet shutdown in 2019, Myanmar’s network disruptions after the 2021 coup, and China’s 2021 mining ban. Node operators independently migrated to censorship-resistant infrastructure without centralized coordination, demonstrating typical decentralized self-organizing adaptability.
Cambridge’s four-layer model (covering Tor relay infrastructure) shows that adding a Tor layer consistently raises the critical failure threshold by 0.02 to 0.10. Even if the clear net is completely removed, Tor nodes can still operate because their relay infrastructure is mainly concentrated in Germany, France, and the Netherlands—countries with extensive submarine cable connections, less affected by peripheral cable failures.
Because Bitcoin nodes are distributed worldwide, and most run over Tor (accounting for 63%), cable cuts typically only impact a small number of nodes in specific regions. Cambridge research shows that in 68 verified cable failure events, 87% caused node changes of less than 5%, far below the threshold needed to impact network consensus. Random cable cuts would need to simultaneously damage 72-92% of transoceanic cables globally to significantly affect Bitcoin.
According to Cambridge’s model, coordinated suppression targeting cloud providers (ASN-directed attacks) can cause noticeable disruptions by removing only 5% of routing capacity, far more efficient than physical cable cuts. The researchers emphasize that this scenario involves policy-level coordinated regulation or service provider shutdowns, not battlefield destruction.
Tor usage grew from nearly zero in 2014 to 63% in 2026, becoming a major part of Bitcoin node infrastructure. It not only provides privacy but also creates an independent structural resilience layer separate from the clear net. Even if clear net nodes are compromised by cloud outages or regulatory actions, Tor nodes can sustain most network operations, greatly enhancing Bitcoin’s overall resilience in extreme scenarios.