TRAI tightens spam rules: When can telcos block callers and disconnect numbers?

The Telecom Regulatory Authority of India (TRAI) has mandated telecom operators to take action against pesky callers or spammers flagged by their AI-based caller ID system, even without the need for consumers filing complaints.

The Telecom Commercial Communications Customer Preference (Third Amendment) Regulations, 2026 introduce a formal definition of "spam" and "spammer", expand the framework to cover application-to-person (A2P) calls and provide for sharing of suspected high-probability spam calling numbers across telecom operators.

Under the amended rules, an A2P call is defined as a voice call initiated by an application, software system or automated platform without direct human dialling, including autodialling, robocalls and prerecorded or artificial voice technologies.

Senders will have to declare in advance to their originating access provider that they intend to use A2P calls, including the Calling Line Identification (CLI) ranges to be used.

The declaration will be recorded on the Distributed Ledger Technology (DLT) platform.

Any A2P call made without such prior declaration will be treated as unsolicited commercial communication and action will be taken against the sender under the regulations.

The originating access providers (OAP) shall, within one business day of the receipt of the flagged calls, identify unique KYC identifiers of the sender associated with such calls and share those details with other telecom operators.

All telecom operators within one day will need to check within their system if there are other connections within their network linked to the suspected spammer in the last 10 days.

"If it is found that five or more CLIs (called line identification) of the Sender have been so identified within the last ten days, all the OAPs concerned shall take action against the Sender," the notification said.

The regulations also provide a mechanism for senders to present their case against such flagging, with the CLI to be unflagged if the justification is found valid.

In case of the first violation triggered using AI system, outgoing services of all telecom resources, including allotted to the sender shall be barred by all the telecom operators for a period of 15 days, irrespective of whether those telecom resources were actually used or not in making such communications.

"For subsequent instances of violations, - (i) all telecom resources, including PRI/SIP trunks, SIMs etc. of the Sender shall be disconnected by all the Access Providers for one year, irrespective of whether those telecom resources were actually used or not in making such communications," the notification said.

In April 2026, AI systems deployed by telecom operators flagged about 771 crore calls and 48 crore SMS as suspected spam, according to data cited by TRAI.

TRAI said the gap between the volume of AI-flagged suspected spam and complaints indicated the need to supplement the complaint-driven mechanism with technological intelligence.

The amended regulations provide for action when complaint data is corroborated with AI/ML-based detection of suspected high-probability UCC activity.

TRAI said this would facilitate earlier identification and investigation of suspected senders while reducing the risk of action based solely on false positives.

The rules also provide for stricter action against senders repeatedly found to be involved in UCC.

For certain repeat violations, telecom resources allocated to a sender, including PRI/SIP trunks and SIMs, can be disconnected across access providers for one year, while the sender can be placed on a blacklist and denied new telecom resources during that period.

However, the violator will still be provided with one telephone number with outgoing services barred. The telephone connection will be allowed to connect with notified emergency outgoing services.

TRAI has also clarified the treatment of service communications. Messages containing OTPs for services such as app login, password reset and two-factor authentication will fall under service messages.

Under the amended rule, subscribers opting to block telemarketing calls from all categories will need to text BLOCK PROMO code to 1909.

"BLOCK PROMO option shall block only promotional types of Commercial Communications for all categories of content, mode, time band and day types but shall not block transaction and service type of Commercial Communications and Government Communications," the notification said.

"All preferences registered under FULLY BLOCK category under present rule will be migrated to BLOCK PROMO category within fifteen days of the notification," it added.

The amended rule will be effective after one month of the gazette notification but in phases over 60-90 days. The norms related to A2P calling will come into effect after 60 days.