“India’s 6G ambitions will mean nothing if carriers keep treating AI as a feature rather than the foundation of network architecture.” — The Mobile Times
India’s AI-ready telecom networks are not a future aspiration — they are a present-tense competitive requirement that most operators are dangerously underprepared for. The country that wants to lead 6G standardisation by 2030 cannot do so by bolting AI onto legacy infrastructure. AI-ready telecom networks demand architectural reinvention, not software patches.
The TMT Position
- India’s DoT has committed Rs 240 crore to indigenous 6G R&D, but zero of that addresses the AI-native core stack carriers actually need
- Reliance Jio and Airtel are testing Open RAN deployments, yet neither has publicly committed to an AI-first radio resource management framework
- ITU’s IMT-2030 specifications require native AI/ML interfaces at every network layer — India’s current 5G rollout satisfies none of those hooks
- Most analysts conflate “AI-assisted” networks with “AI-ready” ones — they are architecturally distinct, and India is building the wrong version
In This Article
Why AI-Ready Telecom Networks Matter More Than Anyone in India Admits
AI-ready telecom networks differ from conventional AI-assisted ones in one irreducible way: the intelligence is embedded in the control plane, not added above it. The ITU’s IMT-2030 framework, finalised in mid-2026, mandates native AI/ML interfaces at the radio access, transport, and core layers simultaneously. India’s current 5G Non-Standalone architecture, which all three major operators deployed at scale, satisfies precisely none of these interface hooks without a full re-architecture of the control plane.
Jio and Airtel have each announced Open RAN trials in 2026, and Vodafone Idea is quietly watching both burn capital on integration. Open RAN matters because it creates the vendor-agnostic layer where AI inference engines can actually sit. But trials are not deployments. South Korea’s SK Telecom has already moved AI-native traffic steering into commercial production across 47 cities. India’s operators are still debating procurement frameworks. That gap will not close by issuing press releases about 6G vision documents.
Is the Infrastructure Gap Overstated? The Counterargument Examined
The reasonable opposing view holds that India’s 5G subscriber base crossed 120 million by Q1 2026, proving the network foundation is sound enough to evolve incrementally. Proponents argue that spectrum allocations in the 26 GHz mmWave band, combined with increased backhaul fibrisation, give operators sufficient headroom to layer AI capabilities progressively. AI-ready telecom networks, they insist, do not require a rip-and-replace cycle. This is a coherent position. It is also incorrect in the specific context of 6G timelines.
Incremental AI layering produces AI-assisted networks, not AI-ready telecom networks. The distinction is not semantic. When network slicing, predictive spectrum allocation, and zero-touch provisioning must operate simultaneously at sub-millisecond latency — which IMT-2030 requires — a control plane that was not designed for AI inference cannot retrofit that capability. GSMA Intelligence’s 2026 operator survey found that 73 percent of carriers who attempted post-deployment AI integration reported latency degradation in at least one critical slice. Incremental does not work at this boundary condition.
What India’s Operators and Regulators Must Do Differently
TRAI needs to stop treating spectrum policy and AI infrastructure policy as separate regulatory domains. AI-ready telecom networks require coordinated decisions: spectrum, power infrastructure, edge compute density, and open interface mandates must move in a single policy envelope. The DoT’s Bharat 6G Vision document, updated in March 2026, reads like a research agenda rather than an operator mandate. It should specify AI/ML interface compliance timelines the way it specifies spectrum release dates. Without that, industry will default to the cheapest compatible option, which is incremental and insufficient.
Success looks measurable. By late 2027, Indian carriers should demonstrate at least two commercial network slices operating under AI-native resource management with zero-touch configuration, as defined by ETSI ZSM standards. C-DoT’s 6G testbed in Bengaluru should produce at least one AI-native RAN prototype that meets IMT-2030 interface specifications, not merely demonstrates machine learning-assisted handover. Tata Communications, which operates significant enterprise network infrastructure, could accelerate this by publishing its own AI-ready network architecture roadmap rather than waiting for operator-led standards to materialise domestically.
The Mobile Times Verdict
India occupies a rare strategic window: it can design AI-ready telecom networks into its 6G architecture from the ground up rather than inheriting technical debt from premature 5G AI retrofits the way Europe and the United States now must. Squandering that window by treating AI as a 5G upgrade cycle rather than a 6G architectural prerequisite would be the single most expensive infrastructure policy mistake of this decade. The ITU clock is running. DoT, TRAI, and India’s three carriers need to treat AI-ready telecom networks as a now problem, not a 2030 problem.
Sources: COAI ↗ | ITU ↗ | DOT ↗ ITU IMT-2030 Framework Specifications (2026); GSMA Intelligence Operator AI Integration Survey Q1 2026; DoT Bharat 6G Vision Document (March 2026); ETSI Zero-Touch Network and Service Management (ZSM) Release 4; TRAI Annual Subscriber Report Q1 2026; SK Telecom AI-Native RAN Deployment Bulletin (2026)
People Also Ask
- What are AI-ready telecom networks and how do they differ from 5G? AI-ready telecom networks embed machine learning directly into the control plane at every network layer. Standard 5G networks add AI above existing architecture. IMT-2030 specifications require native AI interfaces, which current Indian 5G deployments do not natively support.
- When will India launch 6G commercial services? India’s DoT targets 6G commercial deployment around 2030, aligned with ITU’s IMT-2030 global framework. However, without AI-native infrastructure investment beginning now, operators risk missing standardisation windows that South Korea and Japan are already racing toward.
- How can Indian telecom operators prepare for 6G AI requirements? Operators should prioritise Open RAN commercial deployment, adopt ETSI ZSM zero-touch standards, and work with TRAI to align spectrum and edge compute policy. C-DoT testbed outputs must meet IMT-2030 AI interface benchmarks, not just demonstrate assisted automation features.
