Telecom operators now face a sharper question around agentic AI. Should they use it only to cut operating costs? Or can it also drive new revenue?
Cisco argues both goals must move together. At DTW Ignite in Copenhagen, Rana El Desouky Kazamel said service providers are “uniquely positioned” to monetize AI.
Her point is straightforward. Operators already own valuable assets for the AI era. They control power, sites, networks and customer relationships. They also sit close to end users. That matters as AI moves beyond central data centers.
However, new revenue needs a stronger operating base. “You need to build the right foundation with agentic ops efficiency, scaling the networks, operating them in an autonomous manner so that you can deliver services a lot faster, go to market faster,” Kazamel said.
In this view, agentic AI is not just another automation tool. It becomes a platform model for network operations. Operators need a clear target architecture before launching many pilots. Cisco calls this a “North Star architecture.”
That approach can help operators avoid scattered experiments. It also gives teams room to prove value quickly. Early use cases still matter. They can show return on investment and build confidence.
Data remains a large challenge for many carriers. Network data often sits across old systems and separate platforms. Cisco says operators should not wait for perfect data. Kazamel pointed to data retrieval agents as a practical starting point. “We’re going to meet you where the data is,” she said.
Cisco links this thinking to Crosswork AI. The company describes it as a multi-agent framework for network automation. In simple terms, different AI agents handle different operational tasks. They can help find risks, diagnose issues and suggest fixes.
This model could improve network response times. It could also reduce pressure on engineering teams. Yet telecom networks carry high responsibility. A wrong action can affect customers, enterprises and critical services.
That makes governance central to adoption. Engineers need to understand what agents do. Operations leaders also need control over decisions and permissions. Kazamel called change management “the bigger hurdle,” because tools must fit real workflows.
Trust will decide how fast this market moves. Kazamel said AI agents should act like team extensions. “Just like humans, agents have an identity,” she said. “They have a set of guardrails…a set of policies,”
For telecom operators, the message is pragmatic. Autonomous networks will not arrive through disconnected pilots. They need standard platforms, clear policies and reliable data access.
Agentic operations could become a foundation for faster service creation. But carriers must pair ambition with discipline. The winners will likely govern agents as carefully as they deploy them.
