AI traffic growth could strain India's networks as telcos face monetization challenge

Daily multimodal AI use in India is projected to rise from 22% now to 43% by 2030, against a global forecast of 30%, according to Ericsson ConsumerLab research, pushing Indian telecom operators toward targeted uplink, spectrum and backhaul upgrades. Counterpoint Research senior analyst Siddhant Cally said multimodal AI "will introduce new use cases that reshape mobile data consumption, driving not just higher traffic but also changing how it flows," while MTN Consulting's Arun Menon noted India's 5G networks were built for high-volume downlink on mid-band spectrum such as 3.5 GHz, leaving uplink weaker indoors and at cell edges. Ericsson India networks head Nitin Bansal said the networks "are ready for AI but they have to be optimized as per the particular requirements of the specific AI application," as operators weigh 5G-Advanced upgrades amid weak monetization.

Featured Story HPE claims to have beaten Cisco to a proper self-driving network Marvis, described as an AI brain, now unifies all the various parts of the Aruba and Mist portfolios to create a truly autonomous system, says HPE. As AI traffic grows, India's telecom service providers may need targeted network upgrades to handle rising uplink demand even as they struggle with monetization. Daily multimodal AI use in India is projected to rise from 22% now to 43% by 2030, compared with a global forecast of 30%, according to Ericsson ConsumerLab research. As Indian consumers increasingly share images, video and audio with cloud-based AI models, mobile traffic is likely to become more uplink-intensive and less predictable, thus increasing pressure on networks. Traditional communications networks are designed for more downlink traffic because activities such as social media, browsing or video streaming demand downloading more data than they upload. "Multimodal AI will introduce new use cases that reshape mobile data consumption, driving not just higher traffic but also changing how it flows," according to Siddhant Cally, senior analyst at Counterpoint Research. This change in traffic patterns will demand a change in network strategy as well. "India's 5G networks were built primarily for high-volume downlink traffic, using mid-band spectrum, particularly 3.5 GHz. Uplink is weaker indoors and at cell edges because phones transmit at much lower power than base stations. AI applications sending live video, images and audio to cloud-based models will put more pressure on uplink capacity, while on-device processing will generate less additional traffic," explained Arun Menon of MTN Consulting. In addition, AI usage tends to happen in spikes. For instance, a user uploads an image or a voice clip and then waits for a response. "These short sessions can increase radio scheduling demands and latency. Unlike video streaming, where buffering can hide delays, AI interactions require a responsive exchange," Menon added. Existing 4G and 5G networks in India can currently handle AI traffic because it remains a small share of total network traffic, but it is likely to become a challenge in the future. "The networks are ready for AI but they have to be optimized as per the particular requirements of the specific AI application," Nitin Bansal, Managing Director, India and Head-Networks, South East Asia, Oceania and India at Ericsson, told Light Reading in a recent interview. See: Ericsson sees further growth in India's 5G market as AI drives network demand https://www.lightreading.com/wireless/ericsson-sees-further-growth-in-india-s-5g-market-as-ai-drives-network-demand . However, as AI traffic increases, the service providers will need to make upgrades to radio capacity, uplink performance, spectrum and backhaul. "Operators should add sites where demand warrants it, rather than assume a uniform upgrade cycle across India," Menon said. "Delivering consistently responsive experience at scale will require more than incremental radio upgrades. Better uplink performance, increased site fiberization, smarter spectrum utilization will be equally critical," Cally said. The transition to 5G-Advanced 5G-A will also help in better management of the AI-driven traffic. While Reliance Jio, India's largest service provider, has built a 5G-Standalone SA network, Airtel has only recently started transitioning to SA. "The priorities for Indian telcos are uplink capacity, fiber backhaul and completing the move to 5G-SA. 5G-A can improve uplink performance, provide better support for latency-sensitive traffic, support lightweight devices such as AI glasses and wearables, and use AI to reduce radio network energy consumption," Menon said. However, the underlying architecture matters. "5G-A network architecture is likely to deliver limited benefits if the network still relies on a 4G core under NSA or has inadequate backhaul. Microwave-linked towers can also become bottlenecks as radio capacity increases," Menon added. Even as service providers get ready to invest in upgrading networks for AI, they are likely to be acutely aware that growing data traffic may or may not lead to a proportionate increase in revenue. For instance, Telecom Regulatory Authority of India TRAI data for FY26 shows wireless data usage grew 24.74%, while data revenue increased 11.53%. Average industry revenue realization per gigabyte, meanwhile, fell 10.59% from 8.97 Indian rupees US$0.09 to INR8.02 $0.08 . "In India's price-sensitive market, operators will struggle to pass on network investments through higher data tariffs alone. Monetizing premium connectivity and enterprise services, improving networking efficiency, and partnering with AI and cloud players will be key to protecting margins," Cally argued. Read more about: