Artificial intelligence (AI) is no longer an auxiliary capability within telecom. It is increasingly becoming intrinsic to how networks are designed, operated and experienced. The convergence of AI and telecom reflects a structural transition in which networks are evolving from passive connectivity infrastructure into intelligent digital platforms. Telecom networks are emerging as the primary carriers of AI-driven services, while AI itself is becoming the operational intelligence layer embedded within network architecture.
This shift is particularly significant in large-scale digital ecosystems such as India. In a country with over a billion telecom users, the deployment of AI within telecom networks is no longer optional but essential to ensure service quality, operational resilience, energy efficiency and consumer safety. AI is already being deployed for network optimisation, fault prediction, energy management, fraud detection and spam mitigation, demonstrating measurable improvements in both operational efficiency and user protection.
AI-driven network operations and emerging business models
The integration of AI into telecom operations is reshaping both operational economics and long-term business models. From a cost perspective, AI enables the optimisation of capital and operational expenditure through predictive maintenance, intelligent resource allocation and automated network management. Operators are already witnessing efficiency gains in energy consumption, configuration management and bandwidth optimisation through AI-assisted systems.
AI also introduces new revenue pathways. Telecom networks are increasingly being viewed as intelligent service platforms capable of hosting and delivering AI-driven applications at scale. This creates a dual value proposition: enhancing core network efficiency while enabling enterprises, startups and developers to deploy AI-powered services through telecom infrastructure.
At the architectural level, operators face strategic decisions regarding the integration of AI into existing infrastructure. In markets with recent capital investments in 4G and 5G equipment, a full transition to AI-native architecture may not be immediately feasible. As a result, hybrid integration approaches are emerging, where AI capabilities are embedded gradually through bolt-on solutions while preserving existing infrastructure investments. This phased evolution allows networks to modernise without rendering current assets obsolete.
Hybrid intelligence
Historically, AI processing in telecom has been largely cloud-centric, with inference delivered through centralised data centres. However, rising demands for low latency, privacy preservation and personalised services is driving a gradual shift towards more distributed intelligence across network, edge and cloud layers.
Edge intelligence is becoming increasingly critical for real-time responsiveness, privacy-sensitive operations and localised decision-making, particularly for latency-intensive use cases. Meanwhile, cloud systems remain central to large-scale model training, fleet management and complex analytical workloads. Rather than a binary architectural choice, the emerging approach emphasises coexistence and dynamic workload allocation based on performance requirements, data sensitivity and operational efficiency.
In parallel, network-layer intelligence is expected to assume a greater share of automation and optimisation functions. Embedding AI directly into network functions can reduce reliance on distant data centres, improve responsiveness and lower operational complexity. The industry will prioritise automation within the network layer, complemented by selective deployment at the edge and cloud intervention for specialised scenarios.
At the same time, the scale at which AI operates in telecom amplifies its systemic impact. Algorithmic decisions within networks can affect millions of users simultaneously, making trust, transparency and accountability central to AI adoption in this sector. As an essential service infrastructure, telecom networks must ensure that efficiency gains are balanced with consumer rights, explainability and appropriate governance safeguards.
The convergence of AI, connectivity, cloud and devices also necessitates coordinated engagement among telecom service providers, technology developers, regulators, standards bodies and policymakers. Given the complexity of AI-native telecom systems, collaborative ecosystems are essential for managing risks, ensuring interoperability, and supporting secure and responsible innovation across access, core, edge and application layers.
From 5G to 6G
While earlier generations incorporated AI primarily as an optimisation layer, the emerging vision for 6G positions AI as intrinsic to network design rather than an external enhancement. In such architectures, intelligence will be embedded into network configuration, management and evolution.
Current networks already use AI for autonomous configuration and performance optimisation, but the trajectory is moving towards higher levels of autonomy. Industry aspirations include progression from partially automated systems to fully autonomous networks capable of continuous learning from operational data. This evolution will enable networks to self-optimise, self-heal and dynamically adapt to changing traffic patterns, application demands and environmental conditions.
The autonomy in telecom is an incremental journey. Each stage of automation builds on accumulated operational insights, gradually enhancing decision-making capabilities across the network life cycle. In the 6G era, AI-native design is expected to institutionalise autonomy as a baseline feature, transforming networks into intelligent systems capable of managing complexity with minimal manual intervention.
Responsible AI
As AI becomes more deeply embedded in telecom infrastructure, ethical governance and regulatory oversight assume greater importance. Telecom networks interact continuously with citizens, enterprises and public institutions, making responsible AI deployment a matter of public trust. The ethical dimension extends beyond algorithmic accuracy to include transparency, explainability, fairness and accountability in automated decision-making.
Regulatory approaches in India are increasingly aligned with human-centric and risk-based AI governance. Policy initiatives and evolving governance guidelines emphasise safe, accountable and inclusive deployment while supporting innovation. Instruments such as risk-based oversight and regulatory sandboxes support controlled testing of AI-enabled telecom solutions without compromising public interest.
This differentiated approach recognises that not all AI use cases carry the same level of risk. While lower-risk applications may be governed through self-regulation, high-impact deployments affecting consumers require stronger obligations around transparency, explainability and human oversight, particularly as automated intelligence increasingly shapes everyday connectivity experiences.
Fairness in AI-driven network management is another critical operational priority. Automated resource allocation must avoid unintended biases in bandwidth distribution, service quality and network prioritisation across geographies and user segments. Advanced network slicing, assisted management tools and intelligent orchestration mechanisms are therefore being explored to ensure equitable and efficient service delivery.
India’s large subscriber base, extensive mobile broadband reach and rapidly expanding digital infrastructure make a strong case for responsible AI deployment in telecom. Large-scale network operation makes AI-driven automation both impactful and necessary for network efficiency, consumer protection and service quality.
Operational deployments already indicate tangible outcomes, particularly in areas such as spam filtering, fraud mitigation and regulatory oversight, where AI-enabled systems are strengthening network integrity while enhancing consumer safeguards. These use cases demonstrate how responsible AI can simultaneously support operational efficiency and user protection when implemented within structured governance frameworks.
Institutional efforts are also focused on strengthening consumer control and trust in AI-enabled communications, including initiatives related to digital consent management and oversight of commercial communications. Such measures reinforce the principle that AI-led efficiency must be aligned with transparency, accountability and user autonomy.
As telecom networks become more intelligent and compute-intensive, sustainability and security are emerging as key priorities. AI workloads require significant computational resources, making energy-efficient network design critical to long-term scalability. AI-assisted optimisation in network processing and management is already demonstrating measurable improvements in energy efficiency, supporting both performance enhancement and sustainable operations.
Meanwhile, deeper AI integration introduces evolving security dynamics. While intelligent systems improve threat detection and network resilience, they also create new vulnerabilities linked to automated exploitation and higher data exposure. Addressing these risks requires integrated, end-to-end security architectures rather than isolated defensive measures.
Finally, there is a growing shift towards embedding automation directly within network functions to reduce latency, inefficiency and operational complexity associated with excessive reliance on distant data centres. This layered intelligence model, combining network, edge and cloud capabilities, is expected to enhance responsiveness, privacy preservation and operational resilience as telecom networks progress towards more autonomous and AI-native architectures.
The way forward
Looking ahead, the convergence of AI and telecommunications is set to define the trajectory of 6G and future digital infrastructure. Autonomous network capabilities, supported by continuous learning from operational data, are expected to gradually shift telecom systems from assisted management to fully intelligent and self-optimising architectures.
However, technological advancement alone will not determine the success of AI in telecom. Trust, ethical governance and regulatory alignment will remain equally equally. Strong oversight, transparent operational logic and embedded accountability mechanisms will be necessary to ensure that automated intelligence serves public interest while maintaining consumer protection and fairness.
India’s experience in deploying AI in telecom at population scale offers valuable lessons for the global ecosystem, particularly in balancing innovation with inclusion, safety and regulatory prudence. As AI-driven telecom operations expand across borders, issues of interoperability, global standards and ethical alignment will require sustained international cooperation.
Ultimately, the transition towards AI-native telecom networks signals the emergence of trusted, autonomous and inclusive connectivity ecosystems. When designed with responsibility, transparency and collaborative governance at their core, intelligent telecom infrastructure can enhance resilience, strengthen consumer trust, and support the next phase of secure and equitable digital transformation.
Based on a discussion among Pasi Toivanen, SVP, Strategic Government and Industry Initiatives, Nokia; Jagan Shantigram, SVP, Research and Development, Tejas Networks; Dr Vinesh Sukumar, Vice President, Head of GenAI, Agentic AI, Voice AI, Sensing AI, Qualcomm; Erik Ekudden, Group CTO, Ericsson; Anil Kumar Lahoti, Chairman, TRAI; and Ritu Ranjan Mittar, Member, TRAI, at the India AI Impact Summit 2026.
