Navigating the future: AN and AI are shifting APAC's digital landscape

In this Q&A, Henrique Vale (HV) – VP APAC, Cloud and Network Services at Nokia, explores the strategic importance, advances and implications of autonomous operations for the Asia-Pacific region’s vibrant telecom industry. Nokia is a gold partner at Innovate Asia 2025, where it is demonstrating the benefits of AN in a Catalyst proof of concept, Autonomy accelerated: Intent to impact – Phase II.

Why are autonomous networks and AI becoming critical for telecommunication operators in the Asia Pacific (APAC) region today?

HV: APAC's rapidly expanding digital economy, with widespread 5G, IoT and cloud-native infrastructure, demands robust, scalable, agile and efficient networks. Autonomous networks (AN), driven by AI, manage this complexity by enabling zero-touch operations, reducing costs, minimizing errors and accelerating service deployment. AI-driven automation is a strategic imperative for competitiveness and innovation which is crucial in APAC, where rapid urbanization and digital transformation demand seamless, high-quality connectivity and personalized experiences.

What exactly is ‘intent-driven automation’, and how does it transform network operations?

HV: Intent-driven automation shifts from manual configuration to a goal-oriented approach. Operators define high-level ‘intents’ – for example, "ensure premium connectivity for enterprise customers" or "optimize network slice for smart city applications." AI and machine learning algorithms translate them into actionable network policies, automatically configuring, optimizing and self-healing the network. This simplifies operations, reduces manual effort and aligns the network with business goals, even in APAC's multi-vendor, multi-domain environments.

How does closed-loop automation contribute to operational efficiency and customer experience?

HV: Closed-loop automation forms a continuous cycle of observation, analysis, decision-making and execution with minimal human intervention. Network sensors collect real-time data (observe), which AI/ML analyzes for anomalies, predictions, or optimizations (analyze). The system then makes intelligent decisions (decide) and automatically implements corrective actions (act). This real-time responsiveness ensures proactive problem resolution, prevents service degradation and maintains optimal network performance, which translates to higher service availability and consistent quality for APAC customers.

What role does Agentic AI play in achieving higher levels of network autonomy?

HV: Agentic AI represents the next evolution, moving beyond simple automation to intelligent, goal-oriented decision-making. Agentic AI systems understand complex intents, break them into sub-tasks, plan actions, execute and self-correct. Telco agents can make decisions aligned with strategic objectives (for example, adherence to service level agreements or energy efficiency), resulting in fewer outages, faster service, better customer satisfaction and scalable operations in complex networks. Agents learn and adapt, which is invaluable in dynamic networks as it reduces constant human oversight while ensuring high performance.

What are the key challenges operators in APAC face when implementing AN and how can they be overcome?

HV: They face challenges like managing diverse network infrastructures such as legacy alongside 5G and cloud-native, varied regulatory landscapes, and AI/ML skill gaps. Furthermore, proactive security is paramount. For instance, Nokia's NetGuard Cybersecurity Dome, featuring a new GenAI-powered Hunt Assistant, is designed to revolutionize proactive threat discovery in telco networks.

It addresses interconnection attacks within roaming signaling traffic, helping senior security analysts discover new threats and automating anomaly detection for network resilience. Overcoming these challenges requires a phased approach, targeted automation, robust data governance for multi-vendor environments, and continuous upskilling of the workforce. Collaboration with partners and industry bodies like TM Forum also helps standardize approaches and share best practices.

How can data – particularly used with Generative AI – act as a differentiator in building unique and valuable AN applications?

HV: High-quality, real-time data provides essential intelligence for AI models. GenAI amplifies this, enabling deeper insights and new solutions, for example, predicting failures, optimizing resource allocation or generating personalized customer service.

The biggest struggle for CSPs is accessing quality data for AI/GenAI use cases. Nokia Data Suite addresses this, providing the necessary network and subscriber data. It enhances AI applications, including those with an integrated GenAI-assistant, to deliver superior customer experiences. Leveraging operational data with GenAI allows APAC operators to develop bespoke autonomous applications, offering a competitive edge and unlocking new revenue streams.

Can ANs truly operate without human intervention, or is there always ‘a human-in-the-loop’?

HV: While the vision of fully autonomous, ‘lights-out’ networks is compelling, at this stage of the journey, a ‘human-in-the-loop’ approach remains crucial. Until systems achieve complete maturity and can be fully trusted for closed-loop operations across all scenarios, human oversight is essential. Humans are vital for defining strategic intents, overseeing complex decision-making processes, handling unforeseen edge cases, and ensuring ethical considerations are met.

Agentic AI systems are designed to augment human capabilities, taking over routine and complex tasks, but still providing transparency and escalation points for human operators. This collaborative model allows humans to focus on higher-value activities, innovation, and strategic oversight, while the AI handles the operational heavy lifting, leading to a more efficient and resilient network ecosystem.

What tangible business benefits can operators in the APAC region expect from investing in AN solutions?

HV: Benefits are substantial. Operators expect significant OpEx reductions through automated fault management, proactive maintenance, and optimized resource utilization. This improves network performance, service quality, and customer satisfaction – critical for retention in competitive APAC markets. ANs enable faster time-to-market for new services like specialized 5G slices, fostering innovation and new revenue. Industry analysts predict ANs can deliver financial benefits amounting to 5% of CSPs’ annual revenues.

These benefits are demonstrated by initiatives like our Innovate Asia Catalyst, Autonomy accelerated: Intent to impact – Phase II, which showcases Level 4+ AN capabilities in a high-impact B2B context. This Catalyst integrates TM Forum’s intent-based architecture with a closed-loop system, powered by a digital twin, to enables proactive assurance and intelligent automation.

Using Nokia's Digital Operations Center, we can accelerate CSPs’ journey towards fully ANs, leveraging knowledge graphs, digital twins, and GenAI for service orchestration and assurance to support new monetization opportunities, speeding time-to-market by up to 90%, enabling zero-touch assurance, and lowering OpEx, according to Arvinder Khanna, APAC Cloud and Network Services Pre-Sales Leader at Nokia.

How do multi-vendor environments and network slicing fit into the AN vision, especially in diverse APAC markets?

HV: Multi-vendor environments are a reality for most APAC operators. Autonomous network solutions must orchestrate these disparate components seamlessly. Intent-driven automation, with open APIs and standardized interfaces, abstracts vendor complexities, enabling unified control.

Network slicing, a key 5G capability, is linked to autonomy, allowing multiple virtual networks on shared infrastructure, each tailored to service requirements. Autonomous management of slices ensures dynamic resource allocation, performance guarantees, and efficient lifecycle management, crucial for differentiated services across APAC.

What is the outlook for AN and AI in the Asia Pacific region over the next three to five years?

HV: The outlook for autonomous networks and AI in APAC is incredibly promising. Over the next three to five years, we anticipate significant acceleration in adoption. Operators will move towards higher autonomy levels, driven by mature Agentic AI, advanced analytics, and standardization. Focus will shift to orchestrating end-to-end service lifecycles autonomously.

We will see sophisticated use cases emerge, particularly in dynamic network slicing for industry verticals, predictive maintenance, and hyper-personalized customer experiences. APAC is poised to lead in demonstrating how AI-powered autonomous networks drive both operational excellence and innovative service

At Innovate Asia 2025, Arvinder Khanna, APAC Cloud and Network Services Pre-Sale Leader, Nokia, will be participating in a fireside chat at the Level 4+ autonomy: From pilot to production – real progress and results session, which takes place on the Impact Stage on November 25, between 2:30 pm and 4:00 pm.