NEW DELHI, INDIA | October 30, 2026 : India’s telecom regulator has proposed a new framework to govern the use of artificial intelligence across telecom networks and services. The Telecom Regulatory Authority of India (TRAI) has released draft guidelines that could require telecom companies to adopt board-approved AI policies, assess risks associated with AI systems, maintain human oversight and report serious AI-related incidents to the regulator.
The proposal comes as telecom operators increasingly use AI to manage network operations, improve service delivery, detect unusual activity and handle customer interactions. While these technologies can make telecom services more efficient, failures or poorly managed automated decisions could also affect network reliability and consumers.
TRAI has invited comments from stakeholders on the draft guidelines, with November 27, 2026, as the deadline. The proposals are not yet final rules. They will be considered through the consultation process before the regulator decides on the next steps.
What Has TRAI Proposed?
The draft is titled Guidelines on Responsible Use of Artificial Intelligence (AI) in Telecommunication Networks and Provision of Telecommunication Services.
Its central objective is to establish a more consistent approach to AI governance across the telecom sector. The proposed framework seeks to ensure that companies deploying AI systems have clear internal policies, understand potential risks and remain accountable for the outcomes of those systems.
The framework is intended to cover AI used in telecom networks as well as in the delivery of telecommunication services.
Telecom Companies May Need Board-Approved AI Policies
One of the key proposals is that telecom service providers and other authorised entities using AI should develop an internal AI policy.
The policy would need approval from the company’s board of directors or the relevant governing body. It is expected to establish how AI systems are selected, approved, deployed, monitored and managed.
The proposed policy framework covers areas such as:
- Classification and assessment of AI-related risks.
- Approval and monitoring of AI applications.
- Human oversight and intervention procedures.
- Data management and safeguards.
- Handling of AI-related incidents.
- Consumer grievance redressal.
- Accountability for the performance and consequences of AI systems.
The draft also proposes governance and safety committees to oversee the responsible operation of AI systems.
This could make AI governance a formal management responsibility rather than leaving it entirely to technology or operations teams.
AI Systems to Be Classified Into Three Risk Categories
The proposed framework calls for AI systems to be classified according to their potential impact and associated risks.
The three proposed categories are:
1. High-risk AI systems: Applications that could have significant consequences for network operations, consumers, finances or essential services.
2. Medium-risk AI systems: Applications that may create operational or consumer-related risks but have a comparatively more limited potential impact.
3. Low-risk AI systems: Applications with relatively limited consequences if they malfunction or produce an incorrect outcome.
The classification would take into account factors such as the impact on telecom network operations, financial and operational consequences, and the possible effect on consumers.
The purpose of risk classification is to ensure that the level of oversight and safeguards is proportionate to the potential harm.
Human Oversight Will Remain Important
Another important element of the proposal is human supervision of AI systems.
Telecom companies would be expected to establish procedures that allow human intervention when an AI system creates a risk or produces an outcome requiring review.
This is particularly relevant for systems involved in network management, service delivery and customer-related decisions. An automated system may process information quickly, but its decisions could still be inaccurate or inappropriate in certain situations.
The draft also addresses the responsibility of telecom entities when they use AI solutions developed by outside vendors.
Whether a company develops an AI system internally, purchases it from a third party or outsources its operation, the responsibility for the services and outcomes would remain with the relevant telecom entity.
In other words, outsourcing AI technology would not automatically transfer accountability away from the telecom company.
Serious AI-Related Incidents May Have to Be Reported to TRAI
The proposed guidelines also introduce a framework for classifying AI-related incidents according to their severity.
The four proposed levels are:
- Critical
- High
- Moderate
- Low
Under the proposal, incidents classified as critical or high severity would have to be reported to TRAI within the prescribed format, method and timeline.
The framework also proposes a central database to record AI-related incidents and complaints involving consumer harm.
Such records would help companies document what went wrong, investigate the circumstances, take steps to reduce the damage and address consumer complaints.
For customers, this could eventually mean a more structured process for dealing with problems linked to AI-driven telecom services. However, the exact reporting and grievance procedures will depend on the final version of the guidelines.
Why Does the Telecom Sector Need AI Governance?
Telecom networks are no longer limited to voice calls and internet connectivity. They support digital payments, online businesses, government services, emergency communications and other essential activities.
AI can help telecom operators identify network faults, predict equipment failures, manage traffic, detect suspicious activity and improve customer support.
But greater automation also creates new risks. An AI system could make an incorrect assessment, disrupt an operational process, mishandle information or produce a decision that negatively affects a consumer.
In a sector that supports critical digital infrastructure, such failures could have consequences beyond an individual customer interaction.
TRAI’s proposed framework aims to establish clearer responsibilities and safeguards while allowing the industry to continue using AI for innovation and improved services.
How Is the Proposal Linked to India’s AI Governance Guidelines?
The telecom-specific initiative builds on the India AI Governance Guidelines released by the Government of India in February 2026.
The national framework is based on seven guiding principles:
- Trust is the Foundation
- People First
- Innovation over Restraint
- Fairness and Equity
- Accountability
- Understandable by Design
- Safety, Resilience and Sustainability
TRAI’s proposed guidelines seek to apply these principles to the specific requirements of telecom networks and services.
The regulator has said that the consultation process is intended to improve clarity, predictability and accountability in the sector.
What Could Change for Jio, Airtel, Vi and BSNL?
If the framework is finalised along the proposed lines, telecom operators—including Reliance Jio, Bharti Airtel, Vodafone Idea and BSNL—may need to strengthen their internal AI governance arrangements.
This could involve formal approval procedures, risk classification, monitoring mechanisms, human intervention processes and systems for documenting serious incidents.
The impact would depend on the AI applications each operator uses and the final requirements adopted by TRAI.
The proposal should not, however, be interpreted as an immediate instruction to stop using AI or replace existing telecom systems.
What Does This Mean for Mobile and Internet Users?
For ordinary consumers, the proposed changes are primarily about accountability and safeguards rather than an immediate change in mobile plans, recharge prices or internet services.
If the guidelines are adopted, users could eventually benefit from:
- Greater oversight of AI systems used in telecom services.
- More structured handling of complaints involving AI-related problems.
- Clearer accountability when automated systems cause serious disruption or harm.
- Better monitoring of risks associated with AI-driven network operations.
- Human intervention procedures for situations requiring review.
However, the draft does not announce an immediate change to call rates, data tariffs, SIM requirements or existing recharge plans.
Consumers should therefore understand this as a proposed governance framework, not a new set of rules that changes their mobile connections overnight.
How Can Stakeholders Submit Suggestions?
TRAI has invited written comments from stakeholders on the draft guidelines.
Important details:
- Draft released: October 30, 2026
- Last date for comments: November 27, 2026
- Contact person: Akhilesh Kumar Trivedi, Advisor (Networks, Spectrum & Licensing), TRAI
- Email: advmn@trai.gov.in
- Telephone: +91-11-20907758
The draft has been made available through TRAI’s official website. Stakeholders can send their comments electronically, preferably by email.
What Happens Next?
The November 27 deadline will allow telecom operators, technology companies and other stakeholders to submit their views on the proposed framework.
TRAI will consider the feedback as part of the consultation process before deciding whether changes are needed and what form the final guidelines should take.
Until then, the proposed requirements—including board-approved AI policies, risk classification and incident reporting—remain part of the draft framework.
The Bottom Line
TRAI’s proposed AI guidelines could bring a more formal system of accountability to India’s telecom industry. Board-level approval of AI policies, risk-based classification, human oversight and reporting of serious incidents are among the key proposals.
For consumers, the potential benefit lies in stronger safeguards and clearer responsibility when AI-driven telecom systems fail or cause harm. But there is no immediate change to mobile or internet services. The final impact will become clearer after stakeholder consultation and the publication of the regulator’s final framework.
