Making Agentic Operations Happen: WWT's Key Takeaways from Cisco Live 2026

At Cisco Live 2026, one thing stood out above everything else: Cisco is aligning its vast portfolio around a single operating vision that unifies their value proposition and creates a more cohesive, consumable platform. This direction was reinforced throughout the week across keynotes, executive sessions and customer conversations, where a consistent message emerged: the industry has shifted from asking what AI can do to how to operationalize AI at scale.

The implications of this shift are massive. Organizations that architect AI agents well – with the right infrastructure, security models and operational controls – will outpace those that don't. Most of our clients are already experimenting with generative AI, deploying copilots and exploring agentic operations, while emerging capabilities like Mythos are reshaping how systems are governed and controlled. The challenge now is execution: building the infrastructure and operations required to build AI reliably, securely and sustainably.

Major themes from Cisco Live 2026

From AI tools to agentic systems: The network is more important than ever

AI agents are no longer roadmap concepts. From coding assistants like Claude Code and OpenAI Codex to full agentic workflows, organizations are integrating autonomous systems into IT operations today. Rather than simply generating content or surfacing insights, agents are designed to take action, interact with systems and complete tasks within defined guardrails. 

Cisco's vision is centered on building the critical infrastructure for the AI era. The actions taken by AI agents create increased demand. Because agents operate continuously, they are expected to generate significantly more network traffic than human users. As organizations adopt agent-driven workflows, modern infrastructure becomes imperative. Tomorrow's enterprise environments must be built around three priorities: operational simplicity, observability and governance.

Cisco Cloud Control is designed to meet these requirements. It brings together networking, security, observability and agentic operations into a single experience, with AI Canvas providing the generative interface on top. By unifying visibility and management across domains, IT teams can reduce operational complexity while maintaining the governance required to scale AI. The organizations that succeed will adopt an operational model capable of managing this level of continuous system activity.

Zero trust is the foundation in the age of agents

Security was not a standalone track at Cisco Live this year. It was woven into every session and presentation. As AI agents begin acting on behalf of users – accessing systems, initiating MCP connections and making autonomous decisions – perimeter-based models become insufficient. SOC teams must focus on continuous validation of identity, access and intent across both human and non-human actors.

As a result, security assessments must extend beyond AI models to include infrastructure, data and governance. WWT's AI Readiness Model for Operational Resilience (ARMOR) was developed to help clients evaluate security across the full AI lifecycle, from design and deployment through ongoing operations. Applied alongside Cisco's portfolio, our living ARMOR framework provides a structured approach for embedding security and governance before agent development begins.

The reality is that AI security requires visibility and control across far more than a single technology domain. This points to a broader organizational need: a vulnerability operations center with the authority to act across security, infrastructure and end-user computing groups simultaneously. Under this construct, these teams can operate as a unified function capable of defending and responding at machine speed.

Cisco's latest security portfolio supports this model. AI Defense provides visibility into agent behavior and policy enforcement with Cisco Cloud Control, Astrix extends governance to non-human identities and DefenseClaw adds runtime protection by evaluating tools, prompts and connections as they operate. Together, these capabilities will help organizations manage risk across interconnected AI environments.

Cisco Cloud Control with Live Protect

The operating model for agentic infrastructure

Scheduled for General Availability in July 2026, Cisco Cloud Control replaces fragmented IT management platforms with a single control plane for agentic operations. It brings Kubernetes and VM workloads into a unified view, with agents monitoring system behavior, performing root-cause analysis and surfacing recommended actions with confidence scoring. Rather than forcing a binary choice between manual control and full automation, Cisco Cloud Control introduces progressive autonomy for infrastructure operations, allowing teams to begin with human review and gradually expand agent-driven execution as confidence in system behavior increases.

The underlying data layer is powered by Splunk, enabling agents to reason across network, security, application and AI telemetry rather than isolated data sources. Using AI Canvas, teams can query infrastructure using natural language, interact with devices, retrieve insights and manage configurations without switching tools. A digital twin capability enables simulation and validation of changes against a real-time system before production. 

Within this architecture, Cisco Cloud Control brings Cisco-built, partner and customer-developed agents into a shared environment defined by identity, policy and telemetry. This allows automation to scale without fragmenting control, functioning as an agentic harness across infrastructure and operations workflows.

Live Protect extends this approach into vulnerability operations. Instead of waiting for patch cycles that introduce downtime, it applies compensating controls directly in production to reduce exposure while remediation is in progress. Cisco IQ extends visibility into day-to-day operations. Built into Cisco Cloud Control, it provides real-time insight into infrastructure health, inventory and risk, helping teams identify where intervention is needed. Cisco IQ and Live Protect shorten the gap between discovering risk and reducing exposure.

Together, these capabilities eliminate the separation between detection, decision and response. They operate as one system, where infrastructure is continuously observed, evaluated and adjusted. Cisco Cloud Control represents a paradigm shift for IT management – from disconnected tools to coordinated execution.

Top technology trends to watch

1. Right-sizing AI: The case for on-premises workloads

Enabling an AI-native workforce is a strategic priority, but sustaining it is an operational challenge. As organizations increase dependency on cloud-based AI services to scale business operations, tokenomics becomes a critical cost variable. Every inference call, agent interaction and workflow execution contributes to overall spend, but most organizations don't have the visibility or governance to manage it before it becomes material.

The key question is which workloads truly require frontier model capability and which can run more efficiently closer to the data. Not every use case belongs in the cloud. WWT helps organizations operationalize that decision through on-premises AI infrastructure planning and deployment. Some workloads can be run more securely and cost-effectively on-prem, while modular AI factory environments provide a structured path to build these capabilities without starting from scratch. The result is a deliberate placement strategy for AI workloads, optimized for performance and efficiency.

2. Harvest now, decrypt later: Why quantum readiness can't wait

Encrypted data captured today may already be at risk. Adversaries are actively collecting and storing encrypted traffic with the intent to decrypt it once quantum capabilities mature, a strategy known as "harvest now, decrypt later." Q-Day, the point at which quantum computing can break today's encryption standards, is estimated to arrive around 2030 or earlier.

This is not just a future concern – Cisco is treating Quantum Readiness as a current infrastructure requirement. Post-quantum cryptography (PQC) is being embedded across the portfolio as a design principle, including quantum-safe boot and trust mechanisms in infrastructure such as C9000 Smart Switches, which establish a quantum-resistant chain of trust at power-on using NIST-approved algorithms. Across the board, Cisco has committed to extending quantum-safe capabilities across most of its core portfolio by end of 2026.

Through Cisco IQ, this becomes operational. Quantum Readiness Assessments identify where cryptographic exposure exists today, allowing organizations to prioritize remediation based on real infrastructure risk rather than projected timelines. WWT is actively working with clients to incorporate quantum readiness into broader infrastructure modernization efforts, ensuring cryptographic resiliency is built into future-state architectures rather than addressed as a standalone initiative.

The bottom line

The main theme this year was not whether AI will transform the enterprise — it already is. The focus has shifted to how organizations can safely operate autonomous agents across networking, security and applications in real time.

Agentic operations are redefining traditional boundaries between infrastructure, operations, visibility and control. For IT leadership, the question is whether the underlying foundations are strong enough to support continuous, autonomous execution without losing control of cost, risk or performance.

The organizations that succeed will be those that can operationalize intelligence at scale, with infrastructure, security and economics designed for an always-on agentic environment. The payoff is tangible business outcomes: faster operations, lower costs, improved resiliency and better decisions at scale. WWT helps organizations navigate this transition by combining strategy, infrastructure, security and operational expertise to build, validate and scale agentic capabilities responsibly. 

To hear more from our leadership team, tune into WWT's AI Proving Ground Podcast, with Cisco Live episodes coming soon!