Time Will Tell
Why time is becoming the invisible backbone of next-generation telecoms networks
The telecommunications industry finds itself at a strategic inflection point. Communication service providers (CSPs) are being pushed to do two things at once: evolve into platforms for AI infrastructure while transforming their own networks into autonomous, self-managing systems. Much of the industry’s attention is focused on compute, data and connectivity. Yet an older, quieter dependency is becoming newly consequential. Time, long treated as a background utility, is emerging as a constraint on what these ambitions can achieve.
Timing has always been integral to telecoms. It underpins synchronisation, billing and the basic functioning of networks. But as architectures become more distributed and software-defined, the tolerance for imprecision is shrinking. Each new generation of capability, from cloud-native cores to edge-based AI, places tighter demands on accuracy, stability and traceability. What was once assumed is now being scrutinised.
From data to decisions: why AI depends on time
The shift towards AI intensifies this scrutiny. In theory, AI systems depend on data. In practice, they depend just as much on when that data was created, transmitted and acted upon. In distributed environments, where decisions are made across multiple nodes, the ability to reconstruct events accurately depends on consistent timing.
This is where a subtle but important weakness emerges. Timestamps are widely used but rarely verified. Their accuracy is often assumed rather than assured. Yet without confidence in the temporal integrity of data, the outputs of AI systems become harder to interpret, validate and trust.
Improving the precision and traceability of time across networks offers a way forward. Reliable timestamps allow systems to establish causality, correlate events and operate more efficiently. In an AI-driven environment, this is not a marginal gain. It is a prerequisite for trust.
Autonomy raises the stakes
If AI increases the importance of timing, autonomy makes it unavoidable. The transition from centralised, hardware-based networks to distributed, software-defined systems introduces new layers of complexity. Coordination becomes harder. Dependencies multiply. Synchronisation becomes critical.
Autonomous networks rely on the continuous measurement and management of performance. Latency is central to this, but not simply as something to be minimised. What matters increasingly is the ability to measure and control it with precision. Determinism, rather than just latency, becomes the defining metric.
In this context, timing becomes as much a problem of measurement as it is of distribution. Without the ability to measure system behaviour accurately, automation cannot be reliably governed. Precise timing enables that measurement, providing the foundation for confidence, consistency and control in autonomous network operations.
Other industries offer a glimpse of this future. Autonomous transport systems, whether in air, land or sea, depend on precise and resilient timing to maintain situational awareness and operate safely. Telecoms networks, as they move toward similar levels of autonomy, will face comparable constraints. This is illustrated by a recent National Physical Laboratory (NPL) project, delivered in collaboration with Cranfield University and technology partners iQuila and Quantum Dice, which explored autonomous drone operations in environments where GNSS is unreliable or denied. The work showed that alternative, ground-based, traceable time sources can maintain synchronisation and secure communications, enabling safe and continuous operation, and reinforcing the importance of trusted timing in real-world autonomous systems.
The constraints of a single source
These demands are exposing a structural constraint in how time is currently delivered. For decades, telecoms networks have relied heavily on Global Navigation Satellite Systems (GNSS), such as GPS. These systems have provided a convenient and widely available source of precise time. They were not designed, however, to serve as the sole reference for digital infrastructure across industries.
To reduce reliance on GNSS alone, networks must adopt a multi-source approach, complementing satellite signals with additional timing systems to improve resilience and reliability. GNSS signals can be disrupted and as networks become more central to economic activity and public services, the impact of such disruptions becomes increasingly significant.
This has prompted a reassessment. There is increasing interest in diversifying sources of time, including terrestrial alternatives and direct access to national time scales, such as UTC(NPL). The aim is not simply redundancy, but resilience. A system that depends on a single source of truth is, by definition, fragile.
The issue is not only technical. It is also geopolitical. Just as data sovereignty has become a concern in cloud computing, time sovereignty is emerging as a parallel consideration. Knowing where time comes from, and what happens if it is lost, is becoming part of infrastructure strategy.
From constraint to opportunity
What begins as a constraint may yet become an opportunity. As networks evolve towards 5G Advanced and 6G, performance is being redefined. Low latency remains important, but the emphasis is shifting towards predictability. Variability in latency, or ‘jitter’, can be more damaging to user experience than latency itself.
Precise timing is central to managing this transition. While the physical limits of data transmission cannot be overcome, better measurement and synchronisation allow networks to operate closer to those limits, and with greater consistency.
More intriguingly, timing may move from being an internal dependency to a new service for customers. Networks that can generate and distribute precise, traceable time could offer it to other sectors that depend on it, from finance to energy and transport. In doing so, CSPs would not simply consume timing. They would commercialise it.
From invisible utility to strategic capability
For most of its history, timing in telecoms has been invisible. It has worked well enough to be ignored. That is changing. As CSPs pursue AI infrastructure and autonomous operation, timing is becoming both more visible and more consequential.
Time underpins the accuracy of AI systems, the safety of automation, the consistency of performance and the resilience of infrastructure. It also offers a potential source of differentiation in an increasingly competitive landscape.
The industry’s challenge is no longer to recognise that timing matters. It is to treat it accordingly.
Find out more about NPL’s National Timing Centre programme and resilient timing services available to industry: npl.co.uk/ntc
