SUBSCRIBE NOW
IN THIS ISSUE
PIPELINE RESOURCES

Embracing Autonomous Networks
The Road to Level 4 Autonomy

By: Aaron Boasman-Patel

The global telco industry stands at the cusp of transformative change driven by autonomous networks; networks capable of operating, healing, and optimizing themselves with minimal human intervention. For Communications Service Providers (CSPs) looking to both streamline their operations and unlock growth through new services, achieving ‘zero-touch, zero-wait, zero-trouble’ operations is no longer aspirational, it's essential.

Autonomous networking represents a fundamental shift from manual and reactive processes towards predictive, real-time, and self-managing networks. It’s not merely a technological upgrade, it’s a strategic imperative to address rising network complexity, relentless customer expectations, and the need to rapidly launch innovative services.

Level 4 autonomy: crossing the threshold

While network automation itself isn't new, genuine autonomy demands new commitments to both technology adoption and network architecture. In collaboration with our members, TM Forum has developed an autonomy model that benchmarks networks into levels 0 to 5, with Level 4 marking a significant milestone. At Level 4, networks transition from human-dependent automation towards true AI-driven independence. Here, networks autonomously manage, optimize, and self-heal in complex scenarios, continuously learning from their environment.

And global momentum is building. TM Forum members are collaborating on best practices, guidance and implementation aspects toward L4. To date, more than 70 CSPs have committed to achieving Level 4 autonomy by 2027, underscoring industry-wide confidence in the strategic value of autonomous network. Yet, today, most CSPs operate between Levels 2 and 3, indicating significant progress is still required. This gap highlights both the immense potential and the practical challenges CSPs face.

AI and GenAI: Powering Autonomous Intelligence

Artificial Intelligence (AI), particularly Generative AI (GenAI), is the essential enabler of Level 4 autonomy. Unlike traditional automation relying on fixed scripts, AI enables adaptive, real-time decision-making. Machine learning algorithms can analyze vast datasets, predict emerging issues, and proactively manage network resources, drastically reducing manual interventions.

GenAI further extends these capabilities. By interpreting unstructured data, GenAI assists operators through tasks such as real-time troubleshooting, customer interaction, and network optimization. For example, Orange is leveraging GenAI for summarizing trouble tickets, technical documentation retrieval, and natural language querying of network issues, significantly enhancing operational responsiveness.

To fully harness AI and GenAI, networks must embed intelligence into their architecture by design, rather than retroactively integrating it. True autonomy requires intelligent systems capable of making complex, instantaneous decisions across multiple network domains, radio, transport, and cloud without human oversight.

Intent-based, Closed-loop Management

Intent-based management is another key technology that is central to autonomous operations, transforming high-level business goals into real-time network actions. Rather than detailed technical instructions, operators specify desired outcomes or ‘intents,’ and the autonomous network dynamically adjusts itself to meet these objectives.

Coupled with closed-loop automation, intent-based networks continuously monitor performance, analyze data, adapt configurations, and verify outcomes. This proactive loop enables networks to swiftly address issues, often before customers detect them. In practice, this might mean rerouting traffic to avoid congestion or pre-emptively resolving faults, dramatically enhancing customer experience and operational efficiency.

High-value Use Cases Demonstrate Real-world Impact

Real-world examples already illustrate the tangible benefits of autonomous networks. One critical area is autonomous fault management. AI-driven detection and closed-loop resolution significantly lower repair times, improve availability, and minimize customer impact. For instance, Telekom Malaysia boldly guarantees 24-hour broadband fault restoration—a service commitment achievable only through autonomous capabilities.



FEATURED SPONSOR:

Latest Updates





Subscribe to our YouTube Channel