The vision of autonomous networks has been at the centre of telecom transformation for years.
Self-healing systems. Real-time optimisation. Intent-driven operations. Networks that can configure, manage, and optimise themselves with minimal human intervention.
From a technology standpoint, this future is closer than ever.
And yet, most telecom operators are still far from achieving true autonomy.
So what's holding them back?
The Promise of Autonomous Networks
Across the industry, there is strong alignment on the direction of travel.
Standards bodies like TM Forum and 3GPP have defined clear frameworks for autonomous networks, with maturity models outlining the journey from manual operations to fully autonomous environments.
Most European operators today sit somewhere between Level 2 and Level 3 - where partial automation exists, but human intervention is still required for many processes.
The technology required to move forward - AI, advanced analytics, orchestration platforms - is already available and actively being deployed.
So why isn't autonomy scaling?
The Misconception: A Technology Problem
Autonomous networks are often framed as a technology challenge - something that can be solved by investing in AI, automation tools, or next-generation platforms.
In reality, the bigger challenge lies elsewhere.
The bottleneck is no longer technology - it is the operating model that connects technology to value.
Operators are not struggling because the tools don't exist. They are struggling because the environment those tools operate in is not designed for autonomy.
The Real Barriers to Autonomy
1. Fragmented Data Environments
Autonomous systems depend on access to consistent, real-time data across the organisation.
In many telecom environments, data remains fragmented across OSS, BSS, and network layers. Systems operate in silos, making it difficult for AI models and automation platforms to function effectively.
Without unified, governed data, autonomy cannot scale.
2. Legacy OSS/BSS Architectures
Many legacy systems were not designed for the kind of real-time, API-driven interactions required by autonomous networks.
They rely on batch processing, tightly coupled integrations, and rigid workflows. This creates friction when trying to introduce dynamic, event-driven automation.
As a result, even when automation is introduced, it remains limited in scope.
3. Integration Complexity
Telecom environments are inherently complex, involving multiple vendors, systems, and domains.
In the absence of a unified integration layer, organisations rely on point-to-point connections that are difficult to scale and maintain.
This complexity slows down the deployment of new capabilities and limits the ability to orchestrate processes across domains - a key requirement for autonomy.
4. Operating Model and Governance Gaps
True autonomy requires more than just technology - it requires organisational readiness.
This includes:
- Clear ownership of AI-driven decisions
- Governance frameworks for automated actions
- Alignment between business and technology teams
In many cases, these elements are not yet fully established, preventing organisations from moving beyond controlled pilot environments.
Autonomy Is an Operating Model Shift
The path to autonomous networks is not just about implementing new tools - it requires a shift in how telecom organisations are structured and how they operate.
Autonomy demands:
- Seamless coordination across IT and network domains
- Data that is accessible, governed, and contextually enriched
- Systems that can interact through standardised APIs
- Platforms that can orchestrate actions without human intervention
This is fundamentally an integration and operating model challenge, not a purely technological one.
What Needs to Change
Operators that are making progress towards autonomy are focusing on building the right foundations.
Key priorities include:
Composable, API-Led Architecture
Moving away from monolithic systems towards modular, interoperable components that can evolve over time.
Data Readiness
Treating data as a strategic asset - ensuring it is unified, governed, and available in real time across domains.
Intelligent Orchestration
Deploying orchestration platforms that can manage services, resolve issues, and allocate resources autonomously.
Ecosystem Integration
Enabling seamless interaction with partners through standardised APIs, creating a programmable and extensible network environment.
These capabilities enable autonomy to move from concept to reality.
What This Means for Telecom Leaders
For CIOs and technology leaders, the focus should not be on whether to pursue autonomous networks - that decision has already been made.
The real question is whether the organisation is prepared to support autonomy at scale.
Key questions to consider:
- Can systems share data seamlessly across OSS, BSS, and network layers?
- Is there a consistent API layer enabling cross-domain integration?
- Are governance frameworks in place to support automated decision-making?
If the answer to these is no, then autonomy will remain limited to isolated use cases rather than becoming a core operational capability.
Conclusion
The telecom industry is not lacking the technology required to achieve autonomous networks.
The real challenge lies in building the operating model, architecture, and integration capabilities that allow that technology to deliver value.
Until these foundations are in place, autonomy will remain an aspiration rather than an operational reality.
For deeper insights on this topic, download the full report
Frequently asked questions
Autonomous networks are AI-driven systems that can plan, operate, and optimize themselves with minimal human input. They continuously analyze real-time data, correlate events, and take actions aligned with operator intent.
An Autonomous System (AS) is a large, interconnected network or group of networks managed by a single administrative entity (like an ISP, university, or tech company) that shares a unified routing policy. ASs are the "building blocks" of the Internet, communicating via Border Gateway Protocol (BGP).
A Level 4 autonomous network can detect, predict and repair issues autonomously while focusing on self-healing and self-optimization in support of the customer experience. To achieve this the network needs to operate across multiple domains to resolve issues before they affect users.
An autonomous network is a self-managing network infrastructure that uses AI, machine learning, and closed-loop automation to configure, monitor, optimize, and heal itself with little to no human intervention.
Network automation is the use of software tools and scripts to manage, configure, test, and operate physical and virtual network devices automatically, replacing manual CLI-based tasks. It streamlines operations like device provisioning, security updates, and compliance checks, improving network performance, reducing human error, and lowering costs.
|
Pankaj KulkarniSenior Manager Research & InisghtsTorry Harris Integration Solutions |
