Cisco Debuts 'Dialog' Framework to Orchestrate Webex AI Agents

Cisco is updating its Webex cloud collaboration platform with a new agentic framework called Dialog. According to reporting from Engadget, Dialog serves as an "agentic harness" designed to transform fragmented customer interactions into continuous relationships.

Unlike standard chatbots, Cisco claims Dialog allows AI agents to coordinate across backend systems, other agents, and human personnel. These agents are designed to continue working on a customer's behalf even after a specific conversation has concluded. To streamline deployment, Cisco is treating these agents as new teammates, allowing them to ingest operating procedures, knowledge bases, and employee handbooks during onboarding.

From an operational standpoint, the updated Webex platform enables users to invite AI agents into meetings, calls, and spaces to perform multi-step tasks. Engadget notes a specific use case where an agent could pull statistics during a presentation to update a PowerPoint in real-time. To manage safety and data analysis, Cisco is leveraging Splunk, a platform it acquired.

Despite these capabilities, the broader industry faces significant reliability hurdles. Engadget cites research from Carnegie Mellon University indicating that Google's Gemini 2.5 Pro, then the top-performing agent, failed to finish real-world office tasks 70 percent of the time. Other studies suggest failure rates for multi-step tasks range between 60 and 70 percent, as a single error in a complex workflow can lead to cascading failures.