Airtel Kenya AI-Powered Spam Alert Service
Cabinet Secretary, Ministry of ICT & the Digital Economy, Hon. William Kabogo speaks at the launch of Airtel Kenya’s AI-powered Spam Alert Service.

Airtel Africa has revealed that its AI-powered Spam Alert service has detected and flagged more than 205 million spam messages across 13 of its 14 markets in just six months.

Airtel Africa launched the service in May this year to enhance mobile security for its customers. It was designed to protect subscribers from fraudulent and nuisance text messages.  

Available to all subscribers at no extra cost, the service identifies and prefixes the SMS with “SPAM Alert” and provides real-time updates, with the immediate impact being the elimination of the need to download additional applications to manage spam.

According to Airtel, the service has been rolled out in Nigeria, Kenya, Zambia, Uganda, Gabon, Congo-Brazzaville, Malawi, Madagascar, the Democratic Republic of Congo, Rwanda, Tanzania, Chad, and Niger, with Seychelles expected to follow soon.

Of the 205 million flagged messages, Kenya recorded the highest volume with 68 million spam SMS detected, followed by Tanzania with 47 million and Zambia with 33 million.

The company reported that overall spam SMS volumes across its network dropped by 12% since the introduction of the service, with Nigeria recording the most dramatic decline at 84%, a sign that the AI system is helping disrupt spammers’ reach.

Airtel Africa CEO, Sunil Taldar, described the milestone as a major step in ensuring safer digital experiences for mobile subscribers.

“We are proud to pioneer an advanced tech solution powered by AI in tackling spam messages that are a major concern in Africa as smartphone penetration increases. This free service is yet another demonstration of our commitment to consistently innovate to deliver an unmatched experience and safer network to our customers,” he said.

The service uses advanced artificial intelligence and machine learning models to classify messages as spam in real time. Factors such as a sender’s usage patterns, message frequency, and behavioural anomalies are analysed before suspicious messages are tagged.


Leave a Reply

Your email address will not be published. Required fields are marked *