Cisco Introduces AI-First Approach to IT Operations

At its recent Cisco Live 2025 event, Cisco announced AgenticOps, a transformative approach to IT operations that integrates advanced AI capabilities to enhance efficiency and collaboration across network, security, and application domains. Central to this paradigm are the Cisco AI Assistant, Cisco AI Canvas, and the Deep Network Model, each playing a role in redefining modern IT workflows.

AgenticOps: A New Paradigm for IT Operations

AgenticOps represents Cisco's vision for an AI-driven operational model where intelligent agents autonomously manage and optimize IT tasks. This approach leverages real-time telemetry, automation, and deep domain expertise to deliver intelligent, end-to-end actions at machine speed, all while keeping IT teams in control, Cisco said.

Key components of AgenticOps include:

  • Cisco AI Assistant: A conversational interface that identifies issues, diagnoses root causes, and automates complex workflows.
  • Cisco AI Canvas: A shared, interactive workspace featuring a generative UI that unifies data across networking, security, and observability domains.
  • Deep Network Model: A domain-specific large language model (LLM), a specialized LLM trained on Cisco's networking knowledge, including CiscoU and CCIE materials, built on over 40 years of Cisco expertise.

Cisco AI Assistant: Conversational Control Across the Ecosystem

The Cisco AI Assistant serves as the foundational layer of AgenticOps, providing a natural language interface that allows IT professionals to interact with their systems more intuitively. By leveraging the Deep Network Model, the AI Assistant can:

  • Quickly identify critical issues and diagnose root causes.
  • Assist with network configurations and policy management.
  • Automate routine tasks, such as NAC configurations or switch migrations, reducing task time from hours to minutes.
  • Provide actionable insights and recommendations, enhancing decision-making processes.
Cisco AI Assistant
[Click on image for larger view.] Cisco AI Assistant (source: Cisco).

By integrating deeply into Cisco's platform, the AI Assistant ensures that IT teams can easily incorporate its capabilities into their existing workflows, streamlining operations and improving efficiency.

Cisco AI Canvas: A Collaborative Workspace for Cross-Domain IT

Building upon the AI Assistant, the Cisco AI Canvas introduces a generative user interface designed to facilitate real-time collaboration among NetOps, SecOps, and application teams. Key features of the AI Canvas include:

  • Integration of real-time telemetry from sources like Meraki, ThousandEyes, and Splunk, providing a unified view of the IT environment.
  • Dynamic generation of dashboards and visualizations tailored to specific incidents or workflows.
  • Advanced reasoning models that break down troubleshooting into structured steps, guiding teams through diagnostics and remediation.
  • Collaborative features that allow team members to share dashboards, invite collaborators, and persist sessions across time.
  • Execution capabilities that enable the AI Canvas to implement configuration changes or execute runbook steps upon team agreement.

By centralizing data and collaboration tools, the AI Canvas empowers IT teams to resolve issues more efficiently and with greater transparency.

Deep Network Model: The Intelligence Behind AgenticOps

At the core of Cisco's AI innovations lies the Deep Network Model, a specialized LLM trained on over 40 years of Cisco's networking expertise, including CCIE-level content and CiscoU courseware. This model enables:

  • Enhanced understanding of complex network topologies and configurations.
  • Accurate diagnostics and troubleshooting based on real-world reasoning paths.
  • Continuous learning through reinforcement learning techniques and chain-of-thought reasoning.
  • Scalable deployment across various environments, from edge devices to air-gapped industrial IoT systems.

The Deep Network Model's integration into both the AI Assistant and AI Canvas ensures that Cisco's AI solutions are grounded in deep contextual awareness and practical expertise.

Moving to a More Agentic World

Cisco's introduction of AgenticOps, underpinned by the AI Assistant, AI Canvas, and Deep Network Model, marks a significant step forward in the evolution of IT operations. By harnessing the power of AI to automate, collaborate, and optimize, Cisco is enabling organizations to navigate the complexities of modern IT environments with greater agility and confidence, the company indicated.

As Jeetu Patel, Cisco's president and chief product officer, stated in a video announcing Cisco AI Canvas, the company is moving to "a world that's going to be far more agentic, where agents can conduct tasks and jobs almost fully autonomously."

Meanwhile, also at Cisco Live 2025, the company:

  • Introduced a new AI-ready secure network architecture with unified management across Meraki, Catalyst, and new devices.
  • Launched new smart switches and industrial Ethernet switches designed for AI traffic and industrial AI applications.
  • Expanded ThousandEyes vantage points, including new Wi-Fi 7 access points and industrial IoT devices.
  • Announced new 8000 Series Secure Routers with integrated SD-WAN, next-generation firewall, and post-quantum security.
  • Enhanced network security with Live Protect controls and quantum-resistant encryption across multiple layers.
  • Introduced AI-powered Webex Room Vision PTZ camera for intelligent meeting experiences.
  • Expanded Webex AI Agent capabilities for customer service, with pre-built industry templates and enhanced AI engine support.

For more information, visit the Cisco site.

Featured

  • InCommon Academy in action with an Advance CAMP unconference activity at the Internet2 Technology Exchange

    Community-Driven IAM Learning with Internet2's InCommon Academy

    Internet2's InCommon Academy Director Jean Chorazyczewski examines how the academy's community-driven identity and access management learning opportunities support CIOs, IT leaders, and their IAM teams in R&E.

  • businessman juggling cubes

    Anthology Restructures, Focuses on Teaching and Learning Business

    Anthology has announced a strategic restructuring, divesting its Enterprise Operations, Lifecycle Engagement, and Student Success businesses and filing for Chapter 11 bankruptcy in an effort to right-size its finances and focus on its core teaching and learning products.

  • Jasper Halekas, instrument lead for the Analyzer for Cusp Electrons (ACE), checks final calibration. ACE was designed and built at the University of Iowa for the TRACERS mission.

    TRACERS: The University of Iowa Leads NASA-Funded Space Weather Research with Twin Satellites

    Working in tandem, the recently launched TRACERS satellites enable new measurement strategies that will produce significant data for the study of space weather. And as lead institution for the mission, the University of Iowa upholds its long-held value of bringing research collaborations together with academics.

  • Hand holding a stylus over a tablet with futuristic risk management icons

    Why Universities Are Ransomware's Easy Target: Lessons from the 23% Surge

    Academic environments face heightened risk because their collaboration-driven environments are inherently open, making them more susceptible to attack, while the high-value research data they hold makes them an especially attractive target. The question is not if this data will be targeted, but whether universities can defend it swiftly enough against increasingly AI-powered threats.