- The telecom industry is excited about the potential for AI agents
- Multi-agent orchestration will be big in 2026 as AI takes on more CX tasks
- But operators will still need to work on their AI fundamentals
AI is no longer just an experiment for operators, as many are now actively deploying agents for customer service and back-office operations. But where will they go from here ?
Multi-agent orchestration is expected to be a key focus in telco modernization efforts, said Netcracker Director of Strategy John Byrne. Instead of just a single agent solving a problem, he envisions operators will strive to have multiple agents “in service of more broader, bigger outcomes” across billing, CRM and other systems.
“The more you have agents that can serve specific functions, the more you can do things like root cause analysis that looks at a number of different systems and comes back with richer information to make an assessment of what the actual root cause of a network failure might be,” Byrne said.
That being said, he doesn’t think we’ll see a dramatic surge in telecom multi-agent integration. “It’ll be more of an evolution than a revolution.”
As operators move along with AI, Calix VP of Solution and Product Marketing Alan DiCicco predicted 2026 will be the year telcos stop focusing so much on how many connections they have and shift into “communications experience” providers.
Essentially, that means using AI to create “a target segment of one” tailored to each subscriber.
“Imagine a marketer has an initiative, a campaign and an objective that they want to roll out of service,” DiCicco said. “They can use AI to pull subscriber data, demographic data, network information, where the network is, prior service calls.”
To that point, Miguel Alvarez, chief data and AI officer at Orange Business, thinks telecom customer experience will move beyond chatbots and scripted automation and become “every customer’s personal operator.”
Meaning the AI won’t just respond to questions, it’ll be able to handle tasks on a customer’s behalf, such as rebooking flights.
“These intelligent agents will turn support into a seamless, invisible layer of service that anticipates rather than reacts,” he said. “The future of CX isn’t conversational, but operational. It’s AI that does, not just talks.”
Working on AI fundamentals
Optimism for agentic AI is overflowing, but many operators will still need to focus on the foundational work required to enable successful AI adoption, AvidThink Principal Roy Chua told Fierce.
“This includes embracing cloud technologies and agile methodologies, advancing virtualization and making networks more programmable by exposing internal and external APIs — all of which require substantial effort,” he explained.
Operators are in a quandary when it comes to network modernization, according to Byrne. “If we had our way obviously we would have them modernize and update everything all at once.”
They want to upgrade but at the same time, “if you’ve paid for a system that still is only halfway through its expected shelf life, then you’re going to try to keep that system going because it’s essentially paid for,” Byrne noted.
It’s a broken record at this point that disorganized data is the primary obstacle for telcos and their AI aspirations. That doesn’t make a sound data strategy any less important though, as Chua noted operators are working on data collection, filtering, curation, storage and processing.
As for what they’ll do with agentic AI, he thinks operators will deploy agents in “low-risk areas.”
On the OSS side, that will likely involve copilots, troubleshooting, root cause analysis and suggested remediation (the human-in-the-loop component). Regarding billing systems, “innovation will center on agents managing billing, proposing churn reduction strategies and driving proactive customer engagement.”
Chua added work on digital twins and predictive modeling will continue, “with some application of generative and agentic AI in network configuration.”