Telecom giants like AT&T and Verizon are rapidly deploying agentic AI to streamline operations and enhance service delivery. Once experimental, this model—where multiple AI agents handle tasks collaboratively—has become central to telecom AI strategies in 2025.
AT&T is leading the charge with over 410 generative AI agents now in active use. Andy Markus, SVP and Chief AI Officer, said the goal is to break complex problems into smaller tasks, each handled by a dedicated agent. These “agentic workflows” include human oversight to prevent hallucinations and ensure accuracy.
Employees at AT&T are using internal platforms like Ask Operations and Ask Data to automate tasks and interact with datasets using natural language. These modular agents—likened to Lego blocks—can be combined and adapted across use cases, and even integrated with third-party SaaS tools.
Meanwhile, Verizon leverages Google Cloud’s Gemini models to support customer service. The company reports reduced call times and 96% accuracy in AI-assisted interactions. Verizon is developing custom agents and combining them with vendor-provided ones for task-specific use.
Nvidia and Google Cloud are enabling this ecosystem, with new agent-to-agent protocols and AI marketplaces. Industry groups like TM Forum are collaborating to standardize these frameworks, potentially sharing best practices across operators.
While agentic AI is more resource-intensive than a single large model, operators emphasize its higher accuracy and adaptability, especially when paired with their proprietary data.
See What’s Next in Tech With the Fast Forward Newsletter
Tweets From @varindiamag
Nothing to see here - yet
When they Tweet, their Tweets will show up here.
