TRAI is preparing new commercial communication rules to identify legitimate automated business calls as AI voice agents become more common.
The Telecom Regulatory Authority of India (TRAI) is preparing new amendments to India’s commercial communication rules that could bring automated business calls under a more explicit regulatory framework, as companies increasingly use software and AI-powered voice agents to communicate with customers.
Under the amendments being finalised to the Telecom Commercial Communications Customer Preference Regulations (TCCCPR), businesses using automated calling systems will be required to declare such calls to their telecom service providers. TRAI is expected to notify the changes shortly, according to reports citing people familiar with the matter.
The move comes as telecom operators increasingly rely on artificial intelligence and machine-learning systems to detect suspicious calling behaviour. TRAI’s proposed framework is intended to help operators distinguish legitimate automated business communication from spam and fraudulent traffic.
TRAI Wants Automated Calls Declared to Telecom Operators
The central change being considered by TRAI is a requirement for businesses using automated calling systems to disclose that activity to their telecom operators.
This would give telecom networks additional information about the nature of outgoing commercial traffic and help their anti-spam systems distinguish authorised automated calls from potentially abusive calling patterns.
The approach is important because automation itself does not necessarily mean that a call is unwanted. Banks, insurers, delivery companies and other businesses can use automated voice systems for legitimate customer communication, including reminders, confirmations and service updates.
The challenge for telecom operators is identifying the difference between such legitimate traffic and large-scale automated spam.
AI Voice Agents Are Changing the Spam Problem
The issue has become more complicated with the rapid adoption of conversational AI.
Traditional robocalls generally relied on prerecorded messages. Modern AI voice agents can interact with a customer, respond to questions and continue a conversation in real time, making it considerably harder to distinguish an automated call from one handled by a human agent.
Business Standard reported that companies are already deploying voice AI at significant scale. Sarvam, for example, has said its systems handle more than two million voice conversations a day, while Mahindra Finance and Sarvam have reported more than 10 million calls through voice agents across 12 Indian languages.
That creates a new regulatory challenge for India’s telecom networks. A legitimate AI-powered customer-service call and an automated scam can use similar underlying technology, even though their purpose and legality are completely different.
New Framework Could Help Networks Identify Legitimate AI Calls
TRAI’s proposed declaration mechanism could provide telecom operators with another signal when assessing automated traffic.
If a business has properly declared its automated calling activity, operators can potentially use that information alongside existing spam-detection systems and other network-level signals.
The objective is not necessarily to block automated or AI-generated calls. Instead, the emerging framework is intended to ensure that legitimate automated business communication is not incorrectly classified as spam while giving operators stronger tools against misuse.
This distinction is becoming increasingly important as businesses move more customer interactions from traditional call centres to automated voice platforms.
Commercial Calls Must Use Allocated Numbering Resources
The proposed changes also sit alongside TRAI’s existing rules governing numbering resources for commercial communications.
Economic Times reported that businesses and registered telemarketers must use allocated numbering series for commercial communications, while misuse of ordinary 10-digit numbers for such activity can attract regulatory action.
This means the emerging AI-call framework should not be interpreted as a completely new numbering system created specifically for AI.
Instead, TRAI is seeking greater visibility into automated commercial traffic and clearer accountability for the entities using telecom resources to make those calls.
The March 2026 draft amendment already contained provisions dealing with commercial calls made through automated systems. It proposed that an automated commercial call made without prior declaration to the originating access provider could be treated as unsolicited commercial communication.
TRAI Is Already Using AI to Detect Spam
The new measures build on a broader shift in India’s anti-spam system towards network-level AI and machine-learning detection.
TRAI’s draft framework provides for access providers to use AI/ML-based UCC_Detect systems to identify calling-line identities showing behaviour associated with suspected unsolicited commercial communication.
Under the proposed mechanism, a terminating operator can flag a suspected UCC calling-line identity and share the information with the originating operator through the Distributed Ledger Technology platform. Operators can then associate the flagged number with the sender’s KYC information and identify other telecom resources connected to the same sender.
The draft also proposes progressively stronger action when multiple numbers associated with the same sender are repeatedly flagged.
This represents a move away from relying only on individual consumer complaints towards combining complaints with network-level behavioural intelligence.
India Flagged 22.99 Billion Suspected Spam Calls
The scale of the problem explains why TRAI is strengthening its approach.
During the April-June quarter of FY2026-27, AI-based systems used by telecom operators identified 22.99 billion incoming calls as suspected spam, according to government data cited in current reports.
That figure highlights the enormous volume of calling traffic that Indian telecom networks have to process.
At such scale, manual identification is not practical. Automated detection, network intelligence and better information about the source and purpose of commercial traffic are becoming increasingly important for telecom operators.
AI Calls Could Create a New Regulatory Grey Area
The growth of AI voice agents creates a difficult balance for regulators.
If automated calling is treated too aggressively as spam, legitimate businesses could find their customer-service and transactional calls blocked. On the other hand, if automated systems can operate without sufficient identification or accountability, the technology could make large-scale spam and impersonation campaigns easier to conduct.
TRAI’s proposed declaration mechanism attempts to address that middle ground by making the use of automated calling systems more visible to telecom operators.
The approach also places responsibility on the business or entity associated with the telecom resource. According to an official cited by Economic Times, the use of AI solutions for commercial communication is not itself prohibited, but accountability remains with the entity to which the numbering resource has been allocated.
What the New TRAI Rules Could Mean for Consumers
For consumers, the immediate objective is stronger protection against unwanted and potentially fraudulent calls.
A properly identified automated business call could continue to reach customers without being automatically treated as spam, while traffic that violates commercial communication rules could become easier for telecom operators to identify and act against.
The framework could become particularly important as AI-generated voices become more convincing and automated systems are able to conduct increasingly natural conversations.
However, the effectiveness of the system will ultimately depend on how accurately telecom operators can combine declared information with network behaviour, consumer complaints and other spam-detection signals.
TRAI’s AI-Call Rules Are About Accountability, Not Just AI
The proposed changes should therefore be viewed as part of India’s wider effort to modernise its anti-spam framework rather than as a standalone regulation targeting generative AI.
TRAI’s existing draft already addresses automated and Application-to-Person calling, while the latest move brings the issue into sharper focus as businesses increasingly deploy AI voice agents.
If notified, the rules could give telecom operators a clearer way to separate legitimate automated business communication from abusive calling campaigns while placing greater responsibility on businesses using automated systems.
For India’s rapidly developing voice-AI industry, that distinction could be just as important as the anti-spam measures themselves.
