Africa’s biggest mobile network operators are joining forces to address one of the continent’s biggest barriers to artificial intelligence adoption by developing AI systems that understand the languages spoken by millions of Africans.
According to the GSMA’s Mobile Economy Africa 2026 report, six leading telecom operators have launched a continent-wide collaboration to develop artificial intelligence language models trained on African languages, a move expected to broaden access to digital services while positioning Africa’s telecom industry at the centre of the continent’s emerging AI economy.
The initiative, known as the G6 partnership, brings together Airtel, Axian Telecom, Ethio Telecom, MTN, Orange and Vodacom under the banner, “AI language models in Africa, by Africa, for Africa.” First unveiled at Mobile World Congress (MWC) Kigali in October 2025, the collaboration seeks to address one of the structural weaknesses limiting AI adoption across Africa.
Although Africa is home to more than 2,000 languages, representing over 30 percent of the world’s linguistic diversity, today’s leading large language models remain overwhelmingly trained on English and other high-resource languages. As a result, millions of Africans are unable to fully benefit from AI-powered services because the technology often cannot understand or communicate in the languages they use daily.
The report notes that this imbalance extends beyond language preference to digital inclusion and economic opportunity. Without locally relevant language models, many AI-powered services remain inaccessible to people who are not fluent in English or French, restricting the reach of digital innovation across education, healthcare, financial services, agriculture and public service delivery.
“A critical constraint on AI’s reach in Africa is the underrepresentation of African languages in mainstream AI systems,” the GSMA stated in the report.
According to the industry body, developing AI systems that understand African languages could significantly expand access to digital services through voice-based interfaces, allowing users to interact naturally in their native languages while opening new commercial opportunities for businesses deploying AI across the continent.
“The commercial implications are significant,” the report stated, while also noting African language AI enables voice-based interfaces that can extend digital services to users who are not literate in English or French, expand the addressable market for AI-powered products and ensure that the benefits of AI are not concentrated among a linguistically privileged segment of the population.
Rather than focusing solely on language development, the G6 initiative is designed to strengthen the broader foundations required for African-led AI innovation. The partnership brings together telecom operators, researchers, startups and civil society organisations to address four critical pillars identified by the GSMA as essential for sustainable AI development: data, computing infrastructure, talent and enabling policy frameworks.
The report says these investments are intended to ensure that AI systems deployed across Africa’s mobile ecosystem reflect the continent’s linguistic diversity, cultural realities and socioeconomic needs instead of relying entirely on technologies developed elsewhere.
The collaboration has already produced early results. During Mobile World Congress Barcelona 2026, the partners demonstrated what the GSMA describes as the first open Swahili reasoning model. Developed in partnership with MeetKai Zambia, the model can browse online information, translate digital content into Swahili and serve as a blueprint for building similar AI models in other African languages.
Beyond language models, Africa’s mobile operators are increasingly embedding artificial intelligence into their own operations to improve network performance, customer experience and fraud prevention.
The report identifies predictive maintenance, network optimisation and AI-powered customer service among the fastest-growing areas of deployment.
In Uganda, Airtel launched what it described as Africa’s first AI-powered spam alert service in April 2025. Using machine learning algorithms, the system detects and flags fraudulent SMS messages in real time, reflecting growing efforts by operators to combat digital fraud while enhancing customer protection.
Investment is also shifting towards the infrastructure needed to support AI deployment at scale.
In April 2026, MTN participated in a $45 million funding round for ODC, an AI telecommunications startup developing AI-RAN solutions tailored to Africa’s multi-vendor and power-constrained network environments. According to the report, the investment signals an important shift in strategy, with operators increasingly investing in the foundational technologies underpinning AI-enabled mobile networks rather than relying solely on global cloud providers.
The GSMA argues that Africa’s AI ambitions will ultimately depend not only on advanced algorithms but also on strengthening the continent’s AI ecosystem, particularly in areas where significant gaps still exist.
Among the most pressing challenges are limited access to computing capacity, shortages of AI-skilled professionals and insufficient locally generated datasets for training machine learning models.
To close these gaps, ecosystem players are making coordinated investments across the AI value chain.
On computing infrastructure, Cassava Technologies has launched its AI Factory in South Africa, providing GPU-as-a-Service and AI-as-a-Service capabilities to developers, enterprises and mobile network operators. The facility also deploys Nvidia Blueprints to support autonomous network management and improve performance across Africa’s increasingly complex mobile infrastructure.
On the talent front, MTN continues expanding its Skills Academy, a zero-rated digital learning platform available across 16 African markets. The initiative aims to build the technical workforce required to develop, deploy and manage AI systems locally.
Efforts to improve the availability of African language data are also gaining momentum.
In February 2026, Google released WAXAL, an open speech dataset covering 27 sub-Saharan African languages. Developed over three years in partnership with African universities, the dataset provides foundational training material for speech recognition and voice AI applications across languages spoken by more than 100 million people. By making the resource openly accessible, the project is expected to accelerate AI research and product development by African developers and academic institutions.
Attention is equally turning to AI safety and responsible deployment. In March 2026, the GSMA and Zindi launched the African Trust & Safety Large Language Model Challenge to establish a reusable benchmark for evaluating AI models across the continent’s diverse linguistic landscape, helping developers test the reliability and safety of AI systems operating in African languages.
Taken together, the initiatives reflect a broader shift in how Africa’s telecom industry views artificial intelligence. Rather than adopting AI solely as an operational tool, operators are increasingly investing in the foundational assets that will determine how AI is developed, deployed and commercialised across the continent.
The GSMA believes that operators capable of building these capabilities will not only improve operational efficiency but also unlock entirely new markets by making AI-powered services accessible to hundreds of millions of Africans who have historically been excluded by language barriers.
According to the report, expanding access to AI through locally relevant languages and stronger digital infrastructure could play an important role in accelerating productivity and supporting mobile technology’s projected contribution of $290 billion to Africa’s economy by 2030.
As global competition intensifies around artificial intelligence, the report suggests that Africa’s greatest opportunity may not lie in replicating models developed elsewhere, but in building AI systems rooted in the continent’s own languages, cultures and realities, ensuring that the next generation of digital innovation is designed not only for Africa, but increasingly by Africa as well.
