Over the past two decades, the telecom industry has played a pivotal role in driving digital transformation. Now it stands at the threshold of another profound shift as industry leaders look to harness the power of agentic AI to reshape core processes and operations. For telcos that succeed, agentic AI promises to enhance productivity, profitability, and customer service—all while significantly reducing costs.
What is agentic AI?
Agentic AI refers to systems that autonomously perceive, reason, decide, and act to achieve specific goals. These systems possess “agency,” meaning they can operate independently within defined parameters. This makes them far more powerful than traditional rules-based automation, machine learning, or predictive AI, which rely on the analysis of large datasets to estimate probabilities.
Agentic AI combines “sensing” capability with the generative power of large language models (LLMs), enabling it not only to understand and plan but also to act on those plans. Nokia describes this as the "sense, think, act" model:
- Sense: AI agents gather and process data from diverse sources—real-time, historical, structured, and unstructured—to build a contextual understanding of network, service, and business operations.
- Think: An LLM acts as the reasoning engine, interpreting unstructured data, making autonomous decisions, and coordinating with specialized models.
- Act: Agents carry out persona-based tasks (e.g., customer care, security, provisioning), integrating their actions into broader business processes.
This means that agentic AI can unlock the value of the vast—yet often underutilized—datasets telcos possess. And unlike traditional AI systems that require explicit instructions, agentic AI adapts proactively to dynamic environments. This makes it particularly well-suited to the complex, ever-evolving systems within large telecom providers.
Why it matters for telecom
Telecom operators face mounting challenges: massive data volumes, legacy infrastructure, rising customer expectations, and intense financial pressure. Building and maintaining the networks needed for a digital economy demands constant investment, yet regulatory and competitive constraints limit returns. Operational costs consume a significant share of revenue. One recent report estimates that the top 100 global operators spent $1.38 trillion on operations—compared to $1.75 trillion in total revenue in 2024.
Agentic AI offers more than just another technological upgrade—it’s a strategic imperative. To realize its benefits, telecom leaders must understand that this is not merely a technical shift but a chance to rethink business models and redefine how value is delivered.
The next paradigm?
Though still emerging, agentic AI is already showing promise. Research from Appledore predicts it could unlock up to $60 billion in savings by 2030 through reduced operational costs. According to the report, Microsoft believes agentic AI can help telcos improve network resource utilization by 25%.
Generative AI is already enhancing customer satisfaction by enabling real-time, hyper-targeted offers and support. Agentic AI takes this further, automating not just product/service recommendations but also the implementation of changes (with customer consent), improving the customer experience in real time. It augments human roles in customer service and marketing, empowering teams to deliver more value, faster.
A recent TM Forum study highlights agentic AI’s role in helping telcos evolve from connectivity providers to digital service innovators—from “digital telco” to “digital tech co.” Agentic AI is not just about optimizing today’s operations—it’s the foundation for the communications service provider (CSP) of the future.
From vision to action
Momentum is building, and telcos must act quickly. Appledore Research predicts a one hundredfold increase in agentic AI investment in the telecommunications sector over the next five years. In the next article, we’ll explore the key use cases that can deliver quantifiable returns on investment. In the meantime, if you’d like to learn more about how Teradata is already enabling agentic AI in telecom, please get in touch.