# India’s telecom regulator proposes AI rules for operators, from a board-approved policy to incident reports

URL: https://ai-pulse.aglaganova.com/blog/indias-telecom-regulator-proposes-ai-rules-for-operators-from-a-board-approved-policy-to-incident-reports
Published: 2026-11-04
Author: Roman Aglaganov, Full-stack developer
Category: Policy
Region: Asia-Pacific
Tags: regulation, safety

> TRAI’s draft would make telecom companies list every AI system they use, sort them by risk, keep a human in the loop and report serious incidents. Comments are open until 27 November.

![Illustration: a black telecom mast with grey signal arcs, a stack of small grey boxes beside it and one box outlined in red](https://ai-pulse.aglaganova.com/api/media/file/trai-ai-telecom-1.png)

AI is increasingly being built into telecom networks and services, the Telecom Regulatory Authority of India notes. On 30 October TRAI published draft rules on how operators should use it, the first sector-specific application of the India AI Governance Guidelines the government released in February.

## What operators would have to do

The draft would cover any company with a telecom licence or authorisation that uses or plans to use AI, and it makes them responsible for the results whether they build their AI systems or buy them, according to a PTI report on the draft. Its main requirements:

**An inventory:** Every active, inactive and retired AI system must be listed with its source, purpose and risk level, and a system that is not on the list may not be used.

**Risk levels and a policy:** Systems are sorted into high, medium and low risk according to their importance to the network, their financial and operational effects and their possible impact on customers. Each company needs an AI policy approved by its board, covering approvals, incidents, human oversight, data and complaints.

**Oversight:** A senior executive would act as AI Governance and Safety Officer and lead a committee that reviews deployed systems at least once a year and can change or switch them off. The draft says that “All AI systems shall be subject to an appropriate human oversight for risk mitigation.”

**Incidents and checks:** Incidents are graded from critical to low, and critical and high-severity ones must be reported to TRAI. Systems have to be tested for accuracy, robustness, explainability, fairness and drift before they are deployed or changed, and companies must check suppliers before buying AI tools.

## Why it matters

Telecom networks carry emergency calls and payments, so a faulty AI system can do more harm than a bad chatbot answer. The draft is also a test of how India’s light-touch national guidelines, which favour “Innovation over Restraint” as one of their seven principles, turn into concrete duties for one industry. Written comments are invited until 27 November, and TRAI will settle the final rules after the consultation.

## Sources

- [TRAI: Press release 145/2026 on draft guidelines on responsible use of AI in telecommunications (PDF)](https://www.trai.gov.in/sites/default/files/2026-10/PR_No145of2026.pdf)
- [TRAI: Draft guidelines (PDF)](https://www.trai.gov.in/sites/default/files/2026-10/Draft_CP_30102026.pdf)
- [The Week (PTI): TRAI proposes board-approved AI policies, mandatory reporting of serious incidents by telcos](https://www.theweek.in/wire-updates/business/2026/10/30/trai-proposes-board-approved-ai-policies-mandatory-reporting-of-serious-incidents-by-telcos0.html)