Cisco is advancing enterprise AI beyond traditional chatbots with a strategy centered on agentic AI: intelligent software that can act, decide, and coordinate tasks rather than simply assist users. At its WebexOne 2025 conference, the company introduced a suite of AI agents under the “Connected Intelligence” umbrella. These agents are embedded into Webex collaboration and contact center workflows, automating tasks like generating action items, transcribing meetings, suggesting polls, scheduling follow-ups, and even functioning as an AI receptionist for Webex Calling.
The company says their Webex AI Agent provides autonomous customer interaction, resolving inquiries through voice or chat and escalating only when human intervention is required. Cisco AI Assistant enhances human-agent productivity with live transcription, suggested responses, and mid-call summaries, reducing administrative burdens. Multi-agent collaboration is planned for 2026, enabling AI agents to interact securely with third-party systems through standards like agent-to-agent (A2A) and Model Context Protocol (MCP).
Cisco has integrated these AI agents with enterprise platforms including Salesforce Service Cloud, Amazon Lex, and Epic’s EHR software. This allows agents to securely access customer, clinical, and financial data, orchestrating workflows without leaving the Webex ecosystem. Hardware enhancements via RoomOS 26, supported by Nvidia, include AI-powered Director and Notetaker agents for dynamic camera control, audio zoning to reduce distractions, and 3D digital twins of meeting rooms to improve IT planning and optimization.
Beyond collaboration, Cisco is applying agentic AI to network and IT operations. Crosswork Network Automation now incorporates AI agents, while the “AgenticOps” model allows humans and AI agents to jointly troubleshoot infrastructure, monitor performance, and respond to system alerts. This integration reflects a broader strategy to embed AI into enterprise workflows beyond customer-facing services.
Early deployments indicate tangible gains. Certain Webex contact centers report a 31% reduction in operational costs for self-service interactions, and some banking operations have seen an 8% decrease in average handle time. Cisco’s AI Agent Studio centralizes agent development, testing, and deployment, enabling enterprises to accelerate adoption and optimize performance.
With autonomy comes risk. Agentic AI introduces governance and security challenges. Autonomous agents require auditable decision paths, oversight mechanisms, and clear exception handling. Academic research highlights vulnerabilities tied to persistent memory, reasoning errors, and tool integration (ArXiv Paper). Cisco addresses these concerns with network-level protections and tools to detect synthetic media and deepfakes, helping maintain trust in AI-driven collaboration and operations.
Cisco’s AI efforts are occurring in a competitive landscape where rivals like Juniper Networks and hyperscalers such as Microsoft and Amazon are also expanding AI investments. Juniper focuses on AI-driven network automation for enterprise niches, while Microsoft and Amazon integrate AI into cloud platforms, infrastructure, and enterprise productivity tools. Analysts forecast that major technology firms’ AI infrastructure spending will exceed $2.8 trillion by 2029, with roughly $490 billion expected in 2026 alone.
The broader market is trending toward specialized AI models rather than massive, generalized systems. Large language models can cost $4.6–$12 million per training run, incentivizing enterprises to adopt smaller, task-specific models for cost efficiency, compliance, and controllability. Cisco’s approach leverages this trend, balancing scalable AI agents with operational and security oversight.
Despite urgency, adoption remains uneven. Cisco’s 2025 AI Readiness Index reports that 98% of enterprise leaders recognize the importance of AI, but only 4% have reached a “Mature” stage of adoption, highlighting gaps in strategy, governance, talent, and culture. Success will depend on organizations’ ability to integrate AI seamlessly into workflows, monitor agentic activity, and maintain alignment with business objectives.
As AI becomes an integral part of enterprise operations, Cisco’s strategy positions it to compete on multiple fronts: collaboration, IT operations, network management, and security. By embedding agentic AI across its ecosystem, the company aims to make AI an active participant in daily work while addressing governance, integration, and security challenges that define the next era of enterprise productivity.
