The submarine cable connecting two continents looks much as it has for decades – a tube of steel-armoured fibre, laid from a specialised ship, resting on the seabed. Look more closely at what is happening inside that tube, and the picture is rather different.
The cables being designed and deployed today are substantially smarter, more capable, and more situationally aware than anything that existed five years ago.
This technological evolution is not happening in isolation. It is being driven by a confluence of pressures: exploding demand from AI workloads, the specific latency requirements of real-time cloud services, hyperscaler investment in bespoke infrastructure, and perhaps most urgently, a growing concern about physical security in an era of geopolitical friction.
The result is an industry that is innovating faster than at any point in its history, while simultaneously confronting some deeply unglamorous operational constraints.
Cables that can see and hear
The most striking technological development in the sector is the transformation of subsea cables into sensing platforms. Distributed acoustic sensing, which uses the fibre itself as a microphone, detecting vibrations along its entire length, allows operators to monitor cable integrity, detect seismic activity, and identify the presence of ships or other objects in the vicinity of cable routes. Combined with AI-enabled fault detection, this gives network operations centres a degree of situational awareness that was simply not achievable a decade ago.
Valentino Giuseppe at Sparkle articulates why this matters beyond the engineering. Emerging technologies like fibre sensing are transforming submarine cables into real-time monitoring tools, he notes. The cables are no longer passive conduits, they are active participants in their own protection, generating continuous data that operators can use to anticipate faults, identify threats, and respond more quickly when something goes wrong.
This capability is shifting from experimental to operationally essential as network utilisation rates rise and the cost of outages (measured in revenue, regulatory exposure, and reputational damage) increases accordingly. Operators who cannot demonstrate active cable monitoring are increasingly finding that customers, particularly in financial services and hyperscale cloud, are asking pointed questions about their resilience commitments.
The hybrid network blueprint
Alongside the intelligence layer, the physical architecture of global connectivity is being redesigned. The clean distinction between subsea cables and terrestrial fibre networks is dissolving. Next-generation infrastructure is being built as hybrid systems – integrated end-to-end routes that combine transoceanic cable with onshore fibre, designed to reach inland data centre campuses rather than terminating at coastal landing stations and handing off to third-party networks.
Grivner describes the shift in concrete terms. Over the next four years, his organisation’s strategic balance will move towards roughly 60% subsea and 40% terrestrial, compared with the current ratio. The driving logic is not simply cost — it is the need to reach data centre hubs directly, reduce the number of handover points where latency can accumulate, and provide multiple independent paths that reduce exposure to single points of failure.
Ana Nakashidze at AzerTelecom sees the same dynamic playing out across the industry. What is emerging, she says, is the hybrid route – operated as a seamless end-to-end submarine route, with terrestrial elements integrated rather than bolted on. Subsidy models are developing around these routes that could accelerate deployment in markets that have historically struggled to attract investment in connectivity infrastructure.
The repair bottleneck that no one has solved
For all the sophistication of the technology being developed, the subsea sector is confronting a constraint that is neither complex nor novel: there are not enough cable repair ships. The global fleet of specialised cable-laying and repair vessels is finite, ageing in parts, and thoroughly inadequate to handle a sustained increase in fault rates – let alone a scenario involving multiple simultaneous incidents.
Grivner is unambiguous. There simply are not enough ships, he says, and he does not expect the situation to improve meaningfully in the short term. Commissioning and building a cable ship is a multi-year process. The economics of operating one depend on a utilisation model that can shift significantly if repair demand spikes.
Reynolds adds another dimension to the problem. Maintenance capacity and repair logistics are becoming strategic assets, because resilience is ultimately judged operationally – not just architecturally. A network that looks resilient on a diagram but cannot be repaired within a commercially acceptable timeframe is not, in practice, a resilient network.
For data centre executives evaluating connectivity partners, this is perhaps the most counterintuitive lesson from the subsea sector’s current moment of technological sophistication: the most consequential risk may not be a cyberattack on a landing station or a geopolitical cut in the Baltic, but the mundane reality of not having a ship available when one is needed. In an industry racing to build the infrastructure for the AI economy, the humble cable repair vessel has become a strategic chokepoint. That, more than any fibre sensing algorithm or hybrid route architecture, may be the detail that determines whether the digital backbone holds.
Download the report
State of the Sector: Subsea
The findings are drawn from Capacity’s State of the Sector: Subsea report, produced in partnership with senior figures from across the submarine cable ecosystem.
