Telecom network operators face a difficult predicament at the core of their businesses. Networks are more complex, dynamic, and essential for customers than ever before, but the tools and methods used to manage them haven’t kept pace. With the rollout of 5G, the rise of cloud-native architectures, and an explosion of AI enterprise use cases, customers expect telcos to deliver more with less. Industry revenue growth is modest. Operating expenditure pressures are not.

Autonomous networks offer a bold way forward if telcos can realize their full potential. AI-driven decision making, closed-loop automation, and policy-based controls could optimize unit economics at scale, improve reliability, and deliver services at the speed and scale the market demands.

But this is about more than efficiency gains and cost-cutting, and it’s not just a technology story. Autonomous networks are quickly becoming a strategic imperative. They can unlock the agility and intelligence that telcos need to compete in a fast-moving digital landscape, and some are already using the newfound capabilities to address customer problems and gain market share. Autonomous networks could also boost growth by enabling new offerings that telcos previously struggled to scale up because the unit economics didn’t make sense, such as 5G-enabled connected device services (think automatic fall-response systems for the elderly).

Early leaders have quickly made significant investments in autonomous networks, but many telco leadership teams are hesitant or moving cautiously, even if they recognize the value-creation potential. One of the biggest barriers is that the industry has lacked a common way to define or measure how autonomous a network actually is. Without clear metrics or evaluation capabilities, many telcos are struggling to determine where to invest and how much.

The TM Forum created the Autonomous Network Levels Evaluation Tool (ANLET) to help solve this problem. It translates the TM Forum’s autonomy framework—with six levels of autonomy, from fully manual (level 0) to fully autonomous (level 5)—into a practical self-assessment and benchmarking tool that telcos can use to evaluate progress, identify capability gaps, prioritize investments, and align stakeholders around a credible roadmap. TM Forum and Bain are collaborating to link ANLET insights on autonomous networks’ maturity to operators’ financial and operational performance goals, making the capability even more powerful and accessible to the industry.

Since ANLET’s launch in 2024, 45 telcos worldwide have used it to self-assess their autonomous capabilities in at least one type of network “scenario,” such as managing faults in the radio access network (RAN) or optimizing IP network quality. Around 40 of those evaluations by 13 telcos have also been validated by TM Forum and shared publicly. The synthesis of the aggregate results so far—spanning both private and public data—reveals much about where telcos stand in their autonomous networks journey and where they’re headed. Four trends have emerged.

1. Fault management dominates early deployments. About 55% of all assessments thus far have focused on fault management within the core network, RAN, or IP network (see Figure 1). Operators are prioritizing use cases that directly reduce downtime and operating expenditures. These are seen as first-stop priorities—operationally critical, measurable, and often vendor-supported.

Figure 1

In November, several telcos publicly announced high validated scores in fault management. They included China Mobile’s Henan and Guangdong subsidiaries (4.0 scores for IP network fault management), Thailand-based AIS (3.8 in RAN, 3.6 in core network), and Saudi Arabia’s STC (3.6 in IP network). Those numbers dwarf the average score of 2.7 in this category.

2. Telcos are struggling to automate network configuration changes. Among the automation areas tested so far, telcos’ most mature are energy efficiency optimization, service assurance, and network optimization, with average ANLET scores of 3.9 (see Figure 2). (However, it’s worth noting that network optimization has a small sample size.) What’s perhaps more telling is where they’re having the most difficulty with automation: network change management, with a mean autonomy level of just 1.9. This category evaluates a telco’s end-to-end automation of adjustments to network equipment or configuration, from service provisioning to network upgrades.

Figure 2

The low scores for network change management reflect the complexity of legacy operations support systems, especially those involving multiple vendors and network domains, as well as the difficulty of bridging data across organizational boundaries. The lower scores also highlight a challenge that cuts across all autonomous network use cases: Higher levels of autonomy require more than tools; they demand shifts in process, governance, and culture.

3. Asia is leading the way. The region’s large investments in 5G and automation are paying off. ­Telcos in Asia have achieved the highest average ANLET scores on the use cases tested (3.56), ahead of the Middle East and Africa, Europe, and Latin America (see Figure 3). (There were no North America–based telcos in this initial data set.)

Figure 3

The vast majority of the 40-plus validated, publicly shared assessments come from telcos in Asia as well. China Mobile and AIS are among the most active ANLET users. In addition to their high fault management scores, they each have achieved scores of 4.0 in RAN energy efficiency optimization and 3.6 in core network stability, among other validated assessments.

4. There’s a long road ahead, but maturity is trending upward. Executives recognize that the industry remains in the early stages of the shift toward autonomy. The emerging leaders are initially targeting the more easily attainable use cases, and the reality is that most telcos worldwide remain in the level 1–2 range when considering the full scope of autonomy across all network domains and applications.

Nevertheless, ANLET results suggest that autonomy is improving. Between 2024 and 2025, tested use cases’ average maturity levels rose by about 1.1 points to 3.7, with a moderate positive correlation between time and level of autonomy.

Some are making quick progress. For example, AIS announced a validated score of 3.2 for RAN fault management in October 2024; by November 2025, it scored 3.8 in that category.

Beyond the data: The view from the field

In addition to analyzing the existing ANLET data, we interviewed a few leading operators to better understand how they’re using the tool to help transform day-to-day network operations. One instructive example came from an Asia-based telco’s push to embed automation as part of a firmwide transformation. Early efforts have focused on customer complaint handling and incident management.

For customer complaint handling, the team initially focused on automation tools that could improve resolution of issues during the customer’s first call. The team has also tried to increase customer adoption of digital channels, such as a branded app for customer service. The telco has faced technical challenges in consolidating information across functions and departments and then presenting that data in ways that are understandable to customer care agents and customers.

For incident handling, the telco has used network probes to identify affected areas and coordinate field engineers to resolve issues efficiently. This process also focused on reducing the manual analysis workload for the network operations center through automation of alarm handling and rule logic.

This telco is accelerating autonomy through a multipronged approach that includes skills training for staff and codeveloping solutions with vendors. Instead of merely being the recipients of change orders, trusted vendors are sharing with the telco their product roadmaps and more application programming interfaces (APIs) than before. In short, they’re generally acting more like a platform for the telco’s engineers.

ANLET has helped facilitate alignment across the telco’s technical and business functions. In particular, it has given different parts of the business—customer experience, network operations, IT, and vendors—a shared language to talk about automation and help focus everyone on what really matters.

Lessons for telco leaders

Taken together, the ANLET trends and the observations we heard from early-moving telcos point to several practical lessons for executives looking to turn pockets of automation into sustained autonomy that delivers meaningful business benefits.

• Successful efforts have senior executive sponsorship, often at the CEO level. That sponsorship embeds the journey to autonomous networks within the company’s larger strategic priorities or transformation efforts. These executives can help to ensure accountability and break down functional process and data silos that often create barriers to automating more complex processes—where much of the value typically lies.
• Metrics matter, but there is no one-size-fits-all approach to choosing them. Some companies will feel pressure to reduce operating costs, while others will prioritize improving their customer Net Promoter Score^SM or gaining market share. Regardless of which metrics they choose, companies will need to run rapid decision-making processes that prioritize capturing value over traditional “on time, on budget” metrics of many IT efforts.
• Creating a centralized model for scaling AI can unlock progress. Many telcos’ automation and AI efforts are still decentralized, creating platform fragmentation, subscale deployments, and expensive unit economics. Establishing a central model will enable telcos to capture value faster and make it easier to replicate winning use cases across domains.
• Leading operators have an opportunity to influence vendor roadmaps and shape a new ecosystem. If done well, their engineers can wield hardware and software as platforms that can be customized for advantage, rather than black boxes to be monitored.
• Pursuing automation for automation’s sake isn’t the goal. As the industry marches ahead, it will be critical to view every decision through one lens: getting the biggest bang for the buck. The most successful telcos will identify the minimum level of investment and change in automation that will deliver the greatest performance gains.