The Telecom Regulatory Authority of India (TRAI) has issued draft guidelines to govern the responsible use of Artificial Intelligence across telecommunications networks and services, placing enhanced accountability on operators for operational risks, technical failures, and consumer protection. The proposed framework makes human oversight mandatory for automated systems and requires operators to evaluate how machine learning tools impact subscribers, operational stability, and internal commercial processes. Under the plan, service providers must institute board-approved internal governance bodies to track algorithmic deployment and resolve system vulnerabilities before rollout. The regulator has made the consultation draft available on its official portal, setting a deadline of November 27, 2026, for industry stakeholders, technology experts, and the public to submit feedback and suggestions via email.
Mandatory Board-Approved Panels and Three-Tier Risk Classification
Under the draft provisions, telecommunications licensees must constitute an AI Governance and Safety Committee with the formal sanction of their respective boards of directors. Led by senior executives, this internal watchdog will draw representation across diverse functional divisions, including network architecture, customer service, information technology, data administration, and regulatory compliance. The committee will hold statutory responsibility for identifying operational risks, establishing internal controls, and ensuring timely corrective measures. To streamline oversight, TRAI has outlined a three-tier classification model that separates artificial intelligence systems into high-risk, medium-risk, and low-risk brackets based on their potential to disrupt connectivity, compromise sensitive data, or adversely impact users. Deployment protocols will align directly with these categories, requiring internal clearance from designated corporate officials alongside stringent technical audits for high-risk implementations.
Four-Tier Incident Reporting Framework Mandated for Operators
The regulatory draft establishes a structured system for tracking technical breakdowns by classifying artificial intelligence events into four severity levels: critical, major, moderate, and minor. Telecommunications providers will be obligated to maintain comprehensive logs of all operational incidents, covering sudden algorithm malfunctions, unexpected model behavior, and system vulnerabilities that present broader threats to telecommunications services. Companies must conduct thorough internal investigations into every registered event to isolate underlying design defects and prevent recurring failures. For incidents designated as critical or major, operators must report details to the regulator within specified timeframes, enabling official monitors to detect systemic technical vulnerabilities across Indian telecommunications infrastructure.
Centralized Repository to Track Industry-Wide Operational Failures
To strengthen collective digital resilience, TRAI has proposed housing documented artificial intelligence incidents within a central database. The shared registry will compile standardized technical records regarding system breakdowns, algorithmic distortions, and cyber vulnerabilities observed across all participating telecommunications carriers. The consolidated pool of incident data is designed to assist regulators in conducting industry-wide trend analysis, refining threat detection protocols, and supervising the deployment of commercial algorithms. The overarching aim of the proposed regime is to protect subscribers and secure critical communication infrastructure as artificial intelligence adoption deepens across the telecommunications industry. The sector regulator plans to assess all written submissions received before finalising the definitive regulatory guidelines.
