New Cambridge research shows Bitcoin resilience to submarine cable cuts is stronger than many thought. Random faults would need to remove 72%-92% of links to split nodes, while targeted hits on chokepoints or five big hosting providers pose higher risk. TOR usage now boosts physical robustness. Bitcoin has run without pause since 2009, but many still ask what could break it. A new, long-term study from the Cambridge Centre for Alternative Finance offers a data-driven answer. It matches 11 years of peer-to-peer network behavior with 68 confirmed submarine cable faults and thousands of simulations. The headline: it takes a lot to knock Bitcoin offline by accident, and much less if you target the right places. The Cambridge study measures Bitcoin resilience to submarine cable cuts across time and shows how changes like TOR adoption and hosting provider concentration shape the outcome.

Bitcoin resilience to submarine cable cuts: what the data shows

Graceful degradation under random failures

The researchers found that a random wave of cable faults has little bite. In simulations, between 72% and 92% of inter-country submarine cables must fail at once before nodes lose significant global connectivity. That is not how nature usually behaves. They ran 1,000 Monte Carlo simulations per scenario across the full dataset to stress-test this result. The model showed a “graceful degradation” curve, not a sudden cliff. Most of the time, nodes maintain enough paths to talk to each other even when multiple cables are down. Real events back this up. Of the 68 verified cable faults in the dataset, more than 87% caused less than a 5% impact on node connectivity. The most dramatic case came in March 2024, when seabed activity off Côte d’Ivoire damaged 7–8 cables at once. West Africa saw 43% of its regional nodes go dark, but the global impact was tiny: only about 5–7 nodes, roughly 0.03% of the network, blinked out. Markets barely noticed. The study measured the correlation between cable failures and Bitcoin’s price at -0.02. Daily price moves drown out infrastructure noise.

Targeted attacks change the math

Random failures are one thing. Aim at the chokepoints, and the picture shifts. When the researchers removed cables with the highest “betweenness centrality”—the critical routes that connect continents—the threshold to cause similar network damage dropped to about 20%. That is less than one-third of the random-failure burden. The risk is starker at the hosting layer. Today, a meaningful share of reachable nodes sit with a handful of large providers. The study says targeting the top five by node count—Hetzner, OVH, Comcast, Amazon, and Google Cloud—could cause similar disconnection by removing only about 5% of total routing capacity. Unlike storms or anchors at sea, this threat can look like a coordinated policy action or a set of data center outages. When we talk about Bitcoin resilience to submarine cable cuts, this asymmetry is the key insight. Nature has to work very hard to break the network. A focused adversary has a much easier path.

How the network’s structure changed over time

Early strength, mid-cycle concentration, partial recovery

Resilience is not static. The study tracks the network’s changing shape from 2014 to 2025. – In 2014–2017, resilience was highest, with a critical failure threshold near 0.90–0.92. Nodes were more evenly spread across countries and providers. – In 2018–2021, growth came with concentration. Resilience dropped to 0.72 at the low, as many nodes and miners clustered in a few regions. – In 2021, China’s mining ban sparked a broad reshuffle. The network partially recovered to 0.88 in 2022, before settling around 0.78 in 2025. The lesson: geographic and provider diversity raise the bar against both random and targeted shocks. Concentration lowers it.

TOR makes the network harder to isolate

The unexpected boost from hidden locations

By 2025, about 64% of Bitcoin nodes use TOR. That makes many node locations unobservable and pushed some to wonder whether hidden nodes could cluster in fragile places. The Cambridge team built a four-layer model—spanning cables, countries, autonomous systems, and TOR relays—to test this. The result was the opposite of what skeptics feared. TOR relay infrastructure is concentrated in Germany, France, and the Netherlands. These countries sit at the heart of well-connected European cable and land networks. Cutting them off is hard. In the model, adding the TOR layer consistently increased overall resilience versus a clearnet-only baseline, lifting the critical failure threshold by 0.02 to 0.10.

Adaptive self-organization after censorship events

TOR adoption did not rise by central order. It surged after real censorship shocks: Iran’s 2019 internet shutdown, the 2021 military coup in Myanmar, and China’s 2021 mining ban. The community shifted behavior to route around control. That same move also made the physical network tougher to disrupt.

The real threats live at chokepoints

Submarine bottlenecks and cloud concentration

Underwater, a small set of intercontinental cables carry a large share of cross-border traffic. On land, a few hosting providers and network operators serve many public Bitcoin nodes. This creates two kinds of chokepoints: – Cables that bridge continents and have high centrality. – Data center and ISP hubs where many nodes gather for cost and convenience. A storm or an anchor cut is random. A regulator’s order or a coordinated set of maintenance windows is not. That is why the study’s 20% cable-chokepoint and 5% hosting-provider thresholds matter. They map to risks that are feasible for determined actors.

Steps that improve network robustness

No single defender controls Bitcoin’s topology, but many small choices add up: – Run nodes across diverse countries, ISPs, and autonomous systems. – Avoid clustering on a few cloud providers when possible. – Use TOR to reduce exposure to local filtering and make isolation harder. – Support relay diversity by keeping software updated and connecting to a range of peers. – Monitor public metrics on node distribution and hosting concentration to spot drift toward fragility. None of these steps requires central coordination. As the study shows, collective behavior can move the resilience needle.

Why markets barely notice cable cuts

Price reacts to narratives, not to most outages

The dataset puts the price correlation with cable faults near zero. That makes sense. Most cable problems are regional and short-lived. Bitcoin’s peer-to-peer layer can route around them. Meanwhile, price moves on macro news, liquidity, and investor positioning. Even the March 2024 West Africa incident, which was one of the worse multi-cable failures in recent years, barely touched the global node count. It is hard for traders to price a risk that almost never hits the core of the network and that the system can often dodge.

How to assess Bitcoin resilience to submarine cable cuts

What to measure and why it matters

A clear view of risk needs the right yardsticks. The study offers a template: – Map physical links: Track inter-country submarine cables and their known fault history. – Measure centrality: Identify routes and AS-level paths that act as bridges between major regions. – Simulate removal: Use Monte Carlo methods to model random failures, and separate runs for targeted chokepoint removal. – Track providers: Monitor the share of reachable nodes on major hosting companies and ISPs. – Add the overlay: Include TOR relay topology to reflect how much traffic can route through resilient hubs. This approach does not predict the next outage. It estimates how many and which failures would matter, so the community can steer away from fragile patterns.

Context: infrastructure shocks are real, but not destiny

Cable faults happen each year due to natural events and human activity. Political tension can add risk to key straits and landing points. Yet empirical data shows the network has strong buffers against random damage, and it can adapt when incentives change. If the network keeps spreading across borders, providers, and overlays like TOR, those buffers get thicker. In short, the new research draws a bright line between nuisance and danger. Random cuts are nuisance. Intentional action against chokepoints and a handful of large providers is danger. The first is common and survivable. The second is rare but credible, and it is where decentralization still has work to do. Bitcoin resilience to submarine cable cuts looks stronger today than many expected, but not guaranteed. The path to keep it strong is simple: push for diversity, reward openness, and avoid single points of failure. If the network continues to adapt the way it did after past censorship shocks, it will be even harder to isolate tomorrow than it is today.

(Source: https://www.coindesk.com/markets/2026/03/13/bitcoin-can-survive-72-of-the-world-s-submarine-cables-being-cut-but-a-targeted-attack-on-five-hosting-providers-could-cripple-it)

For more news: Click Here

FAQ