TRAI Releases Draft Framework for Responsible Deployment of AI in Telecommunications

TRAI Proposes AI Governance Framework for Telecom

The Telecom Regulatory Authority of India released a consultation paper and draft regulatory framework for the deployment of artificial intelligence systems in telecommunications networks, marking the first sector-specific AI governance proposal by an Indian regulatory authority. The framework covers AI systems used for network optimisation, customer experience management, fraud detection, content moderation on operator platforms, and price discrimination in service bundling.

The central proposal in the framework is a mandatory algorithmic transparency requirement: licensed operators deploying AI systems in network management or subscriber-facing functions must register the AI model's purpose, training data sources, decision criteria, and accuracy metrics with TRAI through a standardised disclosure template. The registration is updated on an annual basis or when a material change to the model occurs.

Independent Audit Regime

The framework proposes establishing a TRAI-empanelled AI Audit Registry, through which licensed operators will be required to commission independent technical audits of their AI systems on a two-year cycle. The audits will examine algorithmic bias — particularly bias in customer service prioritisation or network allocation that disproportionately affects subscribers in rural or low-income demographics — as well as data security practices and adherence to the Digital Personal Data Protection Act requirements.

Operators found to be deploying AI systems with demonstrable discriminatory outcomes in service delivery will face financial penalties under the proposed regime, with penalties scaled to subscriber base size and the severity of the demonstrated bias. The consultation paper explicitly cites instances in other jurisdictions where AI-driven network slicing algorithms inadvertently deprioritised emergency services traffic in lower-income areas as the type of harm the framework aims to prevent.

Industry Response

The Cellular Operators Association of India submitted a preliminary response welcoming the framework's intent while expressing concern about the compliance cost burden for smaller operators, and questioning whether TRAI has the technical expertise to evaluate AI audit reports in-house. The association suggested that TRAI partner with the Indian Institute of Technology system to develop standardised audit methodologies before mandating operator compliance.

Digital rights organisations broadly supported the proposal, calling it a necessary step toward ensuring that AI systems operated by entities with significant market power are subject to public accountability. They called on TRAI to extend the framework to cover AI used in content-related decisions, particularly the use of network-level AI for content blocking and traffic management, which they argued carries significant implications for free expression.