According to the latest “Telco AI: State of the Market” report from GSMA Intelligence, global telecom operators are gradually shifting their AI strategies from improving operational efficiency toward commercial models that can generate direct revenue. The report found that 35% of telco AI deployment initiatives announced over the past six months have explicitly identified revenue targets, indicating that operators are increasingly viewing AI as a new source of revenue growth rather than simply a tool for reducing costs and improving network efficiency.
GSMA Intelligence categorizes the commercial opportunities for telco AI into three tiers. GPU-as-a-Service (GPUaaS), sovereign AI, and vertical industry solutions are regarded as the areas with the greatest growth potential. The second tier includes managed AI services and enhanced enterprise service offerings, while business models such as foundation-model training and open AI markets continue to face greater challenges in achieving commercial viability.
Among these opportunities, AI Factories are emerging as an important component of telcos’ AI strategies. AI Factories are dedicated data centers that integrate high-performance computing, data resources, and AI infrastructure, transforming large-scale data and computing capabilities into services such as automated decision-making, predictive analytics, and AI agents. GSMA Intelligence estimates that AI Factories could potentially generate an approximately 5% increase in revenue for mobile operators, while providing telcos with an opportunity to directly enter the rapidly growing AI infrastructure market.
GSMA Intelligence further notes that even capturing a relatively small share of enterprise demand for AI infrastructure could generate significant new data center revenue for telecom operators. If 5% to 10% of cloud revenue were converted into AI infrastructure services provided directly by telcos, service revenue growth could increase by approximately 1.5 to 3 percentage points. If the conversion rate reached 20%, the potential uplift in service revenue growth could reach approximately 5 percentage points.
Another major market opportunity is sovereign AI. GSMA Intelligence reports that sovereign AI is gradually becoming a mainstream trend across regions. The United States and Canada are currently among the most advanced markets, with approximately 40% of telcos having already launched related products or services. Even in Latin America, where adoption is currently the lowest, around 20% of telcos have introduced some form of commercial sovereign AI service.
Demand for sovereign AI is being driven primarily by highly regulated sectors such as financial services, healthcare, and defense, with particularly strong demand in Europe and the Asia-Pacific region. In the Middle East, Africa, and Latin America, regulatory requirements such as data residency are important drivers of sovereign AI services. Telcos have an inherent advantage in this area because of their local networks, data centers, and regulatory compliance capabilities, enabling them to provide AI services that meet data sovereignty requirements.
However, GSMA Intelligence also warns that telecom operators face a scale disadvantage in capital investment if they seek to expand further into AI infrastructure. Hyperscalers have been increasing AI-related capital expenditure at a rapid pace in recent years. In 2025, for example, Amazon, Google, and Microsoft invested approximately US$92.3 billion, US$54.4 billion, and US$53.0 billion, respectively, in AI-related infrastructure and investment.
By comparison, major Chinese telecom operators collectively invested approximately US$19.6 billion in AI infrastructure in 2025, while telecom operators in Germany and the United Kingdom each invested around US$3.6 billion. The substantial gap in investment scale highlights the challenge facing telcos: while they possess advantages in network infrastructure and local data centers, competing directly with major cloud providers in AI computing and data center markets will require significantly greater capital investment and economies of scale.
Overall, the GSMA Intelligence study indicates that telco AI is moving from an “efficiency improvement” phase toward a "revenue monetization" phase. Going forward, operators’ AI strategies are expected to extend beyond network automation, customer service, and operational optimization into emerging markets such as GPUaaS, AI Factories, sovereign AI, and enterprise AI solutions.
However, the ability to build competitive AI infrastructure with limited capital resources—and to establish differentiated offerings against hyperscalers such as Amazon, Google, and Microsoft—will be critical to determining whether telecom operators can successfully monetize AI and turn it into a sustainable source of future growth.