TRAI Unveils Draft AI Governance Framework for Indian Telecoms, Seeks Public Feedback

Overview

The Telecom Regulatory Authority of India (TRAI) released a draft framework on Friday aimed at governing the use of artificial intelligence (AI) by entities licensed under the Telegraph Act or the Telecommunications Act. The draft is open for public comment until 27 November.

Key Proposals

• Board‑approved AI policy – Every licensed entity must adopt an AI policy that is approved by its board or equivalent governing body.
• Governance structures – Creation of dedicated AI governance and safety committees to oversee risk management.
• Risk classification – AI systems will be grouped into high, medium, or low‑risk categories based on their impact on network operations, financial outcomes, and consumer interests.
• Human oversight – All AI deployments must have a predefined mechanism for human supervision, regardless of whether the technology is built in‑house or sourced from third parties.
• Incident reporting – AI‑related incidents will be classified as critical, high, moderate, or low severity. Reporting of critical and high‑severity incidents to TRAI is mandatory, though the draft does not yet define the exact thresholds or reporting timelines.
• Centralised databases – Telecom operators must maintain a database of AI incidents and consumer complaints, with processes for investigation, mitigation and resolution.

Reported Industry Context

• Major Indian operators such as Reliance Jio and Bharti Airtel already use AI for customer identification, fraud detection and network optimisation.
• Global equipment vendors Nokia and Ericsson are embedding AI into network hardware that Indian carriers are beginning to deploy.

Potential Impact

• Accountability – The framework makes operators responsible for AI outcomes irrespective of the system’s origin (internal development, outsourced, or hybrid).
• Risk mitigation – By mandating board‑level oversight and human‑in‑the‑loop controls, TRAI aims to curb biases, model drift and adversarial attacks that could threaten critical telecom infrastructure.
• Compliance burden – Operators will need to draft policies, set up committees, classify AI assets and maintain incident logs, which could entail additional legal and operational costs.

Next Steps & Verification

1. Public consultation – Stakeholders can submit written comments on TRAI’s portal until 27 Nov. 2024.
2. Final rulemaking – TRAI will review feedback and publish a final version, likely with clearer definitions of severity levels and reporting timelines.
3. Monitoring – Companies should track TRAI notifications and any subsequent circulars for updates on compliance requirements.

How to Verify

• Visit TRAI’s official website (https://www.trai.gov.in) and locate the “Draft AI Governance Framework” under the Consultations section.
• Review the downloadable PDF for the exact wording of each provision.
• Check for any press releases or follow‑up notices that clarify ambiguous points such as incident‑severity thresholds.

--- *The article is based on TRAI’s draft document and publicly available statements. No proprietary quotes or undisclosed data have been added.*