Telcos could be major players for enterprise clients who want fast, customized access to AI services
Telecommunications operators are moving beyond mere pipes. As enterprise demand for AI accelerates, telcos are starting to think about repositioning themselves as AI solution providers, bundling infrastructure, platforms, and productivity tools alongside the connectivity services they’ve long offered.
For telcos, the opportunity is substantial. Traditional connectivity revenue is increasingly commoditized, and AI-as-a-Service represents a critical growth vector. Early adopters in markets like France, South Korea, the UK, India, and Australia are already generating meaningful revenue streams, with significant runway remaining for expansion.
Moving beyond connectivity
Rather than competing solely on network speed and coverage, operators are now positioning themselves as end-to-end AI partners for enterprise clients, marking a pretty substantial shift. This means offering not just bandwidth, but the computational resources, platforms, and expertise required to deploy AI at scale.
Enterprise demand is the primary driver of this shift. Organizations seeking to adopt AI often face prohibitive costs and complexity when they try and develop AI capabilities in-house. Telcos, with their existing relationships and infrastructure, can step in as intermediaries, offering curated solutions that reduce both risk and time-to-deployment. The result is a new category of service that blends connectivity with intelligence.
The pace of change hasn’t been very impressive though. Telcos have long moved slowly, and in the AI era, that could be their downfall.
Revenue models
Telcos are experimenting with a range of approaches to monetize AI capabilities, each targeting different enterprise needs and price points.
AI infrastructure as a service has emerged as a foundational offering. Several operators now provide GPU-as-a-Service, giving enterprises access to computational resources housed in secure, often data-sovereign facilities. This model appeals particularly to organizations with strict compliance requirements or concerns about data leaving national boundaries.
Beyond raw infrastructure, telcos are developing or reselling generative AI platforms tailored to enterprise use cases. Orange, for example, has partnered with French GenAI startup LightOn to launch “Live Intelligence,” a platform combining secure connectivity, private cloud, and orchestration tools. Similarly, Jio’s JioBrain platform and KT’s custom LLM solutions target corporate clients seeking turnkey AI capabilities.
Fine-tuning and customization services represent another revenue stream. Operators like Telus are enabling enterprises to adapt large language models and build custom datasets without requiring specialized AI expertise in-house. This approach lowers the barrier to entry for organizations that want tailored AI without the overhead of building from scratch.
Productivity tools offer a more direct path to enterprise customers. T-Mobile has entered this space with “AI Recaps,” a generative AI-powered note-taking tool sold to business clients. Such offerings demonstrate how telcos can move beyond infrastructure to deliver applications that solve specific workflow problems.
The telco advantage
Telcos bring several structural advantages to the AI-as-a-Service market that differentiate them from pure-play cloud providers and AI vendors.
Infrastructure ownership is perhaps the most significant. Operators already control vast networks, data centers, and computing resources. By leveraging this existing footprint, they can offer AI services at competitive cost structures without the capital outlays required to build from scratch.
Data sovereignty has become a critical differentiator. Enterprises, particularly those in regulated industries, are increasingly concerned about where their data resides and who has access to it. Telcos can position themselves as trustworthy stewards, offering secure, localized infrastructure that keeps sensitive workloads within national boundaries.
Vendor relationships provide another edge. Rather than forcing enterprises to evaluate dozens of AI vendors independently, telcos can curate and pre-vet solutions, reducing complexity and risk for their customers. This curation role adds value beyond what hyperscalers typically offer.
Training AI models from scratch remains expensive; telcos can optimize approaches using open-source tools and pre-built platforms, then tailor solutions to specific enterprise needs.
Key players
Several telcos have emerged as leaders in the AI-as-a-Service space, each pursuing distinct strategies.
Orange has moved aggressively into enterprise AI through its partnership with LightOn. The Live Intelligence platform leverages Orange’s high-bandwidth infrastructure and GPU capabilities housed in French data centers. The company reports that over 90,000 internal users already utilize the platform, providing both a proof point and a testbed for enterprise deployments.
SK Telecom operates at significant scale, with a personal AI agent reaching users in South Korea. The company has also established dedicated divisions for enterprise AI transformation services (AIX) and AI datacenters (AIDC), and launched GPU-as-a-Service offerings. A partnership with Perplexity aims to develop a global personal AI agent, extending SK Telecom’s reach beyond its home market.
Swisscom has partnered with Nvidia to build GenAI supercomputers intended for both internal use and enterprise clients. The collaboration positions Swisscom as a provider of high-performance AI infrastructure in the European market.
Challenges ahead
Despite the opportunity, telcos face meaningful obstacles in scaling their AI ambitions.
Organizational barriers could be the biggest. Many operators must break down internal silos and refactor legacy platforms to support enterprise-wide AI deployments rather than isolated pilot projects. This transformation requires both cultural and technical change. Telcos aren’t known for moving all that quickly — giving others an opportunity to swoop in.
The talent gap presents another challenge. AI expertise is in high demand, and telcos aren’t used to paying the money required to compete with hyperscalers and AI startups for that talent. And, there’s more direct competition from hyperscalers. AWS, Azure, and Google Cloud have massive AI capabilities and deep customer relationships. Telcos will need to differentiate through connectivity integration and data sovereignty propositions rather than trying to match hyperscaler scale directly.
The market remains in its early stages, with considerable runway for expansion as more organizations move beyond pilot projects to production deployments. Those telcos that can successfully navigate the organizational, talent, and competitive challenges stand to capture a significant share of enterprise AI spending in the years ahead.
