Artificial Intelligence

Cisco Pushes AgenticOps Vision to Help IT Teams Manage AI-Era Networks

At Cisco Live 2026, executives outline a future where artificial intelligence-driven agents monitor infrastructure and recommend corrective actions.

Mickey McCarter

Mickey McCarter is the senior editor of StateTech magazine.

As artificial intelligence agents begin taking on more work inside enterprise environments, Cisco is introducing a new operational model designed to help IT teams manage increasingly complex networks at machine speed.

At Cisco Live 2026, Cisco executives unveiled a series of networking, security and observability enhancements built around what the company calls “AgenticOps,” a framework that combines AI agents, automation and human oversight to identify, diagnose and resolve IT issues.

Joe Vaccaro, Cisco senior vice president of network platforms and ThousandEyes, tells BizTech that AgenticOps represents a significant evolution beyond traditional AIOps platforms.

“Traditional AIOps helped humans identify anomalies and pinpoint where a problem might exist,” Vaccaro says. “But it still kept the human at the center of decision-making. With AgenticOps, we’re utilizing agents as digital teammates to extend the scale of what teams can operate.”

One of Cisco’s key announcements was the introduction of agentic actions for networking through Cisco Cloud Control, a new management platform that brings together networking, security, observability and collaboration tools under a single interface.

Vaccaro describes the approach as an “agentic loop” that enables organizations to sense issues, diagnose root causes through AI-driven reasoning, remediate problems, validate changes and ultimately deploy fixes.

“Now we’re identifying not only here’s the event, but here’s why it occurred, here’s a suggested action, and, if you want, we can take that action automatically on your behalf,” he says.

Click the banner below to keep up with Cisco Live coverage.

Building Trust in AI-Powered Operations

Cisco executives acknowledged that organizations may be hesitant to allow AI agents to make changes to production environments.

To address those concerns, Cisco is emphasizing transparency, validation and governance throughout its AgenticOps strategy.

AI-generated recommendations will include reasoning logs that show how the system reached a conclusion, giving IT teams visibility into the evidence behind each recommendation, Vaccaro says.

The company is also introducing digital twin capabilities that allow proposed network changes to be tested before they are deployed.

“Statistics show that the majority of network disruptions are human-caused,” Vaccaro says. “The validation step becomes a critical component.”

Using a digital twin, organizations can simulate configuration changes, evaluate their impact and verify outcomes before applying them to live environments.

“It allows operators to write in pencil before they commit in permanent marker,” Vaccaro says.

Cisco is also pairing AI-driven recommendations with predefined workflows and runbooks designed to provide deterministic execution, rather than allowing agents to make arbitrary changes on their own.

AI Drives New Demands on Networks

Cisco executives framed the AgenticOps strategy as a response to broader shifts in enterprise infrastructure driven by AI.

During his Cisco Live keynote address Tuesday in Las Vegas, Cisco Chair and CEO Chuck Robbins argued that AI systems are increasing the importance of networking infrastructure rather than diminishing it.

“The network is truly more powerful than the node,” Robbins said, noting that AI models, applications and agents become significantly more valuable when connected together.

Robbins said Cisco expects AI-related network traffic to triple within the next three years as organizations deploy more AI workloads, robotics and autonomous systems.

Jeetu Patel, Cisco’s president and chief product officer, echoed that message during the keynote, describing a future in which AI agents operate continuously and generate substantially more network traffic than human users.

“Humans click, but agents swarm,” Patel said.

Cisco research indicates that an AI agent generates roughly 450% more network traffic than a human performing the same task, Patel said.

Those demands are driving Cisco’s investments in networking infrastructure, security and observability, as well as the company’s efforts to simplify operations through Cisco Cloud Control.

For Vaccaro, however, the long-term goal is not replacing IT professionals but helping them keep pace with increasingly complex environments.

Cisco IT has already begun using early versions of the company’s AgenticOps capabilities, including “ambient agents” that continuously monitor environments, identify issues and propose fixes through Webex channels.

“We’re not replacing the need for humans,” Vaccaro says. “But we’re giving these IT operations teams the scale that they need to operate in this fast-paced, highly complex environment.”

Pekic/Getty Images