Looking ahead to the 2030 network

In the paper, we ask you to imagine this network of the future, where AI weaves through an adaptive network. The scenario illustrates how AI-driven autonomous networks enable seamless, intelligent connectivity across dynamic environments. Vehicles, drones, satellites and terrestrial networks operate as a coordinated, layered system, with AI agents predicting needs, provisioning resources and reverting configurations automatically once demand passes. Connectivity extends beyond simple access, supporting applications such as emergency response, autonomous navigation, environmental monitoring and real-time AI analysis via satellite backhaul.

This vision highlights a fundamental shift in telecoms from static, best-effort connectivity to intelligent, adaptative networks integrated across diverse technologies. And, crucially, it is set in 2030 – a mere four years away. This is not the distant future – in fact, it is already happening.

The physical network is becoming more dynamic. Satellite services are being integrated by default, drones can provision temporary capacity in emergencies, and private 5G networks are increasingly common across major enterprise deployments. The network is no longer a single, static infrastructure – it is a growing, heterogeneous ecosystem of terrestrial, aerial and space-based assets.

This is why AI is more than just an upgrade, but a system-level shift that will redefine telecoms.

The implications of this shift are profound. AI is not transforming one layer of the telecoms stack at a time, but reshaping the entire system simultaneously.

This breadth is what gives AI its power, but it is also what makes it difficult to act on. When AI affects everything, incremental experimentation and isolated pilots are no longer enough. Decisions made in one domain inevitably shape others. The choices operators make today will determine which path their networks follow tomorrow.

So if this is what networks will look like in 2030, what must operators do today to make it real?

Innovation built on strategy

Translating broad vision into a clear, executable strategy and technology plan requires explicit choices about where intelligence lives, how decisions are made and how automation scales safely across the network.

This is the strategic challenge we help telecom leaders solve: translating AI ambition into a clear, executable strategy and technology plan that can be delivered at speed with a deep tech approach.

So where to start? In our view, three ideas work as practical planning lenses for operators looking to move from experimentation to execution.

1. A hierarchy of intelligence

Telecoms networks are inherently hierarchical, spanning the core, access, edge, home and device. When approached deliberately, this structure becomes a strategic advantage rather than a constraint.

Our paper calls this the ‘hierarchy of intelligence’, where intelligence ranges from centralised brain models to distributed, localised agents. A coherent AI strategy defines what intelligence belongs at each layer, balancing latency, privacy, resilience and cost. Decisions requiring millisecond response times belong close to the edge or device, while broader optimisation and learning can be handled centrally.

Making these choices explicitly helps avoid fragmented architectures to unlock differentiated capabilities that are hard to copy. Telecoms operators need to move fast to become leaders in edge AI and hierarchical intelligence or risk opening the door to new competitors.

2. Autonomy as a strategic foundation

Autonomy is often framed as a future ambition. In practice, it is the foundation that enables progress today. AI-driven autonomy allows networks to sense conditions, make decisions and act with minimal human intervention. The near-term benefits are lower operating costs, improved performance and more consistent service quality.

Our paper with Verizon frames this transition in terms of TM Forum autonomy levels, where the majority of operators are planning on progressing from Level 3 ‘conditional autonomy’ to Level 4 ‘highly autonomous networks’ by 2030 to deliver lower OPEX and faster time to market. We support operators in identifying where autonomy can deliver fast, low-risk returns, creating momentum while laying the groundwork for broader transformation.

3. Designing for multi-altitude networks

Future networks will not be defined by a single access technology. Satellites, High Altitude Platforms (HAPs), drones, non-terrestrial networks, private networks and dynamically deployed capacity will become part of a single, unified system.

AI is the orchestration layer that makes this complexity manageable. Planning for a multi-altitude future requires rethinking coverage, resilience and scale – not as a static infrastructure, but as a dynamically composed network. These considerations have direct implications for architecture, partnerships and operating models, reinforcing the need for joined-up planning.

A practical roadmap to turn AI strategy into execution

Once these strategic lenses are in place, the challenge becomes how to move from strategy to execution. A common pitfall is attempting to do too much at once – instead, you can tactically sequence AI adoption across three horizons to gain value as you progress:

Today: Apply AI to optimise networks and operations, stabilising margins and reducing cost. This is where autonomy begins to deliver measurable ROI, through anomaly detection, automated network change and service assurance improvements.

Tomorrow: Use those gains to turn existing assets, such as the edge or home gateway, into platforms for differentiated services. AI becomes the foundation for new products and revenue streams rather than just an efficiency tool.

Beyond: Prepare for an agent-driven world where service provider agents interact with user agents to dynamically conceive of, provision and deliver services on demand. In this world, the trust relationship with the user, proximity and edge-intelligence determine the responsiveness (and therefore, relevance) of the network.

Engage today, win tomorrow

The transition to AI-powered networks is not optional, but the role operators play in that future is still being decided. What sets this moment apart is not only the scale of change, but its speed. Many of the capabilities described in the CC/Verizon paper are already emerging through early deployments of autonomous operations, multi-altitude networks and agentic AI. The architectural choices made now will determine whether operators become intelligent service orchestrators or increasingly commoditised connectivity providers.

Waiting for certainty is not a neutral stance – it is a strategic decision to surrender value and control to hyperscalers, platforms or new entrants.

But the future is not pre-written, and for now, operators still have the opportunity to choose their path. Those that act with intent, grounded in a clear strategy and practical technology plan, can capture value where intelligence is created, shape how AI is embedded into networks and define the next era of connectivity rather than simply supporting it. As the deep tech powerhouse of Capgemini, CC can help industry leaders bridge the gap between vision and execution – from strategy and architecture to prototyping and deployment.

The window to lead is narrow, but it is still open. If you want to shape the future of telecoms, reach out to continue the conversation.