Beyond Automation: How agentic AI is redefining the telecom landscape

Agentic AI is more than just an evolution of AI. It represents a fundamental shift in how telecom operators deliver services as it enables AI systems to reason, plan and act independently, with intent, personality and context-awareness. For telecom, agentic AI represents the next leap, a move from assistive tools to goal-driven agents that transform both operations and engagement.

India is uniquely positioned to march ahead in the agentic AI revolution, supported by a robust digital infrastructure, a thriving AI ecosystem and a skilled workforce. According to Deloitte’s “State of GenAI” report, over 80 per cent of Indian businesses are exploring the development of autonomous agentic AI systems, and 70 per cent are aiming to leverage GenAI for automation.

A leap beyond traditional automation

Unlike reactive assistants (which are being used mostly now), these agents understand objectives, operate across multiple environments, and dynamically collaborate with humans and systems. In telecom, this shift to intelligent agents unlocks new possibilities in customer care, network performance and service delivery. Agentic AI does not just make existing workflows faster; it redesigns them entirely.

Verticalising AI for telcos

Telcos are traditionally composed of siloed domains, network, customer service, billing and customer relationship manager that rarely share data in real time. To unlock the full potential of agentic AI, a unified data architecture is essential.

While the telecom sector has embraced GenAI, many deployments fall short of true operational integration. A telco-grade agent, one that is verticalised, is deeply embedded with telco-specific skills, ontologies and reasoning capabilities. These agents go beyond surface-level interactions, delivering network-aware, context-sensitive responses and executing intelligent decision-making rooted in domain expertise.

For AI agents to operate effectively in the telecommunications industry, they must understand the language, logic and structure of telco systems. This is where ontology, a formal representation of domain knowledge that enables consistent data interpretation across platforms and tasks, becomes necessary. A well-defined telco ontology provides AI agents with a contextual understanding of service plans, technical specifications, billing structures and customer interactions. To support this, a robust ingestion pipeline is needed, capable of handling a variety of data types that agents rely on, including real-time streams, time-series data, and other structured and unstructured sources.

At Amdocs, our foundational effort has been to unify these domains to provide real-time, contextual insights. In a proof of concept with a major North American telecom operator, we deployed agentic AI as an assistive layer in customer care. The results were remarkable: call handling time dropped by 63 per cent, first-call resolution improved by 50 per cent and transactional Net Promoter Score (tNPS) increased by the same margin.

Digital Twins and AI-Orchestrated Engagement

While AI has long supported telecom networks for fault detection, predictive maintenance and optimisation, its role in customer engagement has been limited. A cornerstone of this transformation is the concept of digital twins, which are real-time virtual replicas of a physical network. By providing a scalable and risk-free environment for training and testing, these simulations allow agentic AI to master complex decision-making, anticipate outcomes and adapt its strategies, all without disrupting real-world operations. The result? A truly autonomous network that can self-optimise, dynamically manage resources and even self-heal, delivering enhanced performance, reduced operational costs and an elevated customer experience.Unlike traditional bots, which are bound by scripted responses, agentic AI brings the ability to reason, plan and act based on a user’s context, enabling adaptive, emotionally intelligent interactions tailored to individual needs. Success in this new era will not come from deploying generic bots; it will come from investing in agentic AI that connects, adapts and elevates the human experience. The future of engagement is here, and it is not human-led; it is AI-orchestrated with human intent at its core.

Strategic enabler across domains

Agentic AI is more than a customer care tool – it is a strategic enabler across the telecom value chain. Its benefits include proactive maintenance, dynamic network optimisation, seamless automation.

Challenges and the road ahead

Adopting Agentic AI comes with hurdles. Many telcos still operate on legacy systems that do not support real-time data sharing. These must be modernised to provide the context AI agents need. Governance is another critical factor. AI must operate transparently and ethically. Trust in AI agents will be earned only through consistent, fair decision-making. Moreover, telecom workforces need to evolve. As AI takes over repetitive tasks, employees must shift toward roles focused on oversight, strategy and human-centric service. Yet, the opportunities far outweigh the challenges. Agentic AI is not just a technology – it forces telcos to reimagine their systems, their roles and their relationships with customers.