For years, telecom operators have monetised connectivity through data plans, voice services, roaming and enterprise contracts. Artificial intelligence is beginning to introduce a different commercial possibility: selling the infrastructure, capabilities and services that businesses need to build and operate AI applications.
The challenge is that investing in AI infrastructure does not automatically create a profitable business.
Recent developments suggest that operators are exploring several ways to participate in this emerging market. Singtel, for example, has introduced a token-as-a-service offering through its RE sovereign cloud business, reflecting a growing interest in commercial models tied to AI workloads rather than connectivity alone. Telecom industry discussions are also increasingly focused on enterprise AI, cloud infrastructure and network APIs as potential sources of new revenue. <Cite refs={[“turn377445search18”,“turn377445news8”]}/>
Connectivity Alone May No Longer Be the Whole Product
An enterprise deploying AI may need reliable connectivity, secure data movement, cloud infrastructure, low-latency access and predictable performance. In some cases, it may also need network capabilities that can be requested programmatically by an application.
Telecom operators already control important parts of this infrastructure. The opportunity is to package those capabilities into services that solve identifiable business problems.
Network APIs are one example. Through standardised interfaces, developers can access capabilities such as number verification, device information and quality-on-demand services. The GSMA’s 2026 Open Gateway research identifies growing interest in API aggregation, which allows developers to access capabilities across multiple operators through a common integration point. <Cite refs={[“turn377445search1”,“turn377445search5”]}/>
However, technical availability is only one part of the equation. Operators must still determine who will buy these services, how they should be priced and which partners will distribute them.
A network capability that developers can access easily is useful. A capability that solves a measurable business problem and has a sustainable commercial model is a product.
The Operational Layer Will Decide How Quickly Operators Can Move
New services often introduce complexity that is less visible than the technology itself.
A telecom business launching an AI-enabled enterprise offering may need to manage new service plans, partner agreements, usage records, billing rules and customer entitlements. If those functions remain disconnected, each new offering can create another set of manual processes and integration requirements.
This is where modern BSS and OSS architecture becomes strategically important. API-driven integration, flexible charging models and automated service provisioning can help operators introduce new commercial offerings without treating every launch as a separate systems project.
For MVNOs and MVNAs, the same issue applies at a different scale. They may not own the underlying network infrastructure, but they still need to manage subscriber relationships, service configuration, billing and partner dependencies. Platforms such as TelcoEdge operate within this broader shift toward more programmable telecom operations.
The Hard Question Is Commercial, Not Technical
Telecom operators have infrastructure, enterprise relationships and extensive operational experience. Technology companies bring developer ecosystems, cloud platforms and AI expertise. Neither side necessarily controls the entire customer experience.
That creates an important strategic choice. Should operators focus on supplying the infrastructure behind AI services, or should they build higher-value offerings that sit closer to the enterprise customer?
There may not be one answer for every operator. But the businesses that succeed will likely be those that connect their technical capabilities to a clear customer need, rather than assuming that every new technology creates a new revenue stream.
What do you think will become the stronger long-term opportunity for telecom operators: providing the infrastructure that powers AI, or selling AI-enabled services directly to enterprises?