The Frontier Telco: Why AI-native transformation is telecoms’ next competitive imperative

CSPs are uniquely positioned to become trusted enablers of the AI economy, but fragmented adoption and legacy operating models are holding many back. Upendra Kohli, Executive Vice President and Industry Head for Communications Service Providers at Infosys, outlines why the window for meaningful transformation is narrowing.

‎TM Forum: You have coined the term “Frontier Telco”, framing it as a concept that telcos should be adopting now. Can you briefly explain what you mean here?

UK: The Frontier Telco concept emerges from a clear-eyed diagnosis of where the telecom industry stands today. The core telco business has become commoditized, and growth has become a challenge. In the absence of significant revenue growth, the margins are continuously diluted due to stiff competition. This is compounded by the broader geopolitical and technological changes driven by AI, which could give higher leverage to hyperscalers and in future to space technology.

The advent of AI has opened a new front for telco and created an exciting monetization opportunity by addressing multiple other challenges such as sovereign, responsible, and scalable guardians of the AI revolution. However, the adoption has been constrained by the current enterprise setup and capital.

The Frontier Telco concept addresses these constraints and opens an opportunity to expand and deliver comprehensive enterprise transformation to create an accelerated AI-first organization. This will require mindset shift for colleagues and customers to deliver business value through innovation. This concept details how a telco can navigate to a more autonomous operating model, identifying specifically which elements of the business can be optimized through AI, to deliver enhanced customer value with strategic oversight with human in the loop.

The risk of not transforming is far greater than the risk of acting. If telcos don’t make this shift, they’ll stay trapped in legacy cost structures, fall behind on innovation and customer service, and risk becoming utilities in the AI economy instead of value creators. The Frontier Telco mindset is therefore not a vision for the distant future: it is a call to act now, before the window of opportunity closes.

TM Forum: What would differentiate a “frontier” telco from a more traditional telecoms organization?

UK: Several things set a Frontier Telco apart:

Operating model structure: The Frontier Telco is structured around three interconnected layers: strategic steering and enablement; customer value orchestration; integrated intelligent operations. Each of these layers are human-led and AI-first.

Customer engagement model: Frontier Telcos replace fragmented account structures with permanent, cross-functional customer value stream teams, aligned to specific segments. Such teams are accountable for delivering end-to-end (E2E) outcomes. They focus keenly on business metrics such as customer churn, time to market, customer lifetime value, and NPS.

Workforce and talent: Once AI begins to optimize ~90% of operational work, the human workforce must shift from doing the work to designing, supervising, and elevating the work which requires entirely new roles such as AI orchestrators, AI ethicists, and value designers that simply do not exist in most traditional telcos today.

Task elimination: In the integrated intelligent operations layer, things are highly optimized, with AI executing close to 90% of tasks, acting as assistant, executor, and eventually autonomous operator, with humans validating only exceptions. Traditional telcos, by contrast, still have most operational work performed manually or with fragmented, piecemeal robotic process automation.

TM Forum: Two of the TM Forum’s three missions are centered on AI and Data, and Autonomous Networks. How have these two missions informed your own approach?

UK: Yes, these two missions are critical to create a future vision for a telco and formulate the Frontier Telco operating model.

The Frontier Telco concept delves deeply into these two areas. The AI & Data Mission’s call to break down silos and become AI-native enterprise is visible in how Frontier Telco treats data. The transition from data silos to context-driven enterprises capitalizes on big data as a strategic asset, exposed through governed APIs and continuously curated by AI, making data the central nervous system of the enterprise, not a bottleneck.

For autonomous networks, underlying data components, sources and the inherent quality need to be addressed and move towards AI-native architecture to achieve zero touch fulfilment and assurance. This will also enable integrated intelligent operations to realize fully autonomous self-healing networks, moving towards TM Forum AN Level 4 autonomy.

The Frontier Telco concept goes further than either mission in isolation and connects the two, by treating network autonomy and enterprise AI as parallel workstreams. The concept is structured around the three interconnected layers outlined above, where the gains from autonomous network operations feed directly into elevated human performance at the customer and strategy layers.

TM Forum: Given that both AI and AN are set to feature heavily at DTW Ignite, how will Infosys stand out from the crowd?

UK: A combination of structured operating model, proprietary technology fabric, broad partner ecosystem, and human transformation focus is what differentiates Infosys at any industry forum.

We offer a holistic framework, not point solutions. Rather than offering individual AI tools, Infosys brings the Frontier Telco blueprint which converges structured, end-to-end architecture spanning operations, customer engagement, and strategic governance.

Infosys Topaz Fabric is the technology backbone to deliver this blueprint. This is a multi-layer service suite that unifies infrastructure, models, data, applications, and workflows into a composable, agent-ready ecosystem.

We also work within a strong ecosystem, supported by partnerships with leading players such as Nvidia, OpenAI, Anthropic, ServiceNow, Palo Alto Networks, Cognition, Salesforce, and hyperscalers such as Google, AWS, Microsoft, and Oracle.

Finally, instead of focusing on technology deployment, Infosys frames this as a talent transformation as well, requiring reskilling at every level. This is what makes it competitive and a leading proponent of the Frontier Telco concept.