Industry executives and experts share their predictions for 2026. Read them in this 18th annual VMblog.com series exclusive.
By John Capobianco, Head of AI & Developer Relations, Itential
For more than a decade, network automation has been “the future.” Scripts replaced manual changes, playbooks promised consistency, and infrastructure-as-code brought discipline to previously artisanal workflows. Over the last three years, artificial intelligence accelerated that momentum dramatically. In 2026, however, the industry will cross a more consequential threshold. Networking will move from fragmented automation efforts into agent-driven, orchestrated systems that fundamentally change how networks are designed, operated, and staffed.
Prediction 1: MCP Was the Bridge: Agents Are the Destination
The adoption of Model Context Protocol (MCP) marked a critical turning point for AI in networking. Prior to MCP, large language models operated largely as isolated assistants – useful, but disconnected from enterprise tools, telemetry, and intent. MCP normalized how context is shared between models, tools, and systems, creating a common interface for real-world interaction.
With that foundation now established, the industry is ready to move into Agent Development Kits (ADKs) and Agent-to-Agent (A2A) protocols. In 2026, AI agents will no longer be novelty chat interfaces bolted onto tools. They will be first-class actors in network operations, capable of reasoning, delegating tasks, coordinating with other agents, and executing changes through controlled workflows.
Instead of a single monolithic “AI,” organizations will deploy swarms of specialized agents: agents that understand routing intent, agents that reason over telemetry, agents that validate change risk, and agents that coordinate remediation. A2A communication will allow these agents to negotiate actions, validate assumptions, and escalate intelligently – without human intervention for every decision.
Prediction 2: Orchestration Becomes the Center of Gravity
The early days of network automation were a necessary but messy phase. Scripts lived on laptops, playbooks were duplicated across teams, and jobs were scheduled in isolation. Automation existed, but orchestration was optional.
That era is ending.
In 2026, orchestration will be the backbone of serious network automation programs. As AI agents become capable of initiating actions, the need for centralized coordination, policy enforcement, and lifecycle control becomes non-negotiable. Orchestration platforms, like Itential, will provide the guardrails: sequencing, approvals, rollback logic, observability, and compliance.
This shift mirrors what happened in application development years ago. Individual scripts gave way to pipelines; pipelines gave way to platforms. Networking is following the same maturation curve. The “wild west” of automation – where every team reinvents workflows and nothing is reusable – will be replaced by shared, governed, and extensible orchestration frameworks.
Prediction 3: Agent-Driven Operations Replace Static Runbooks
Runbooks have always been a compromise. They document what humans should do when something goes wrong, but they rarely reflect the full complexity of modern networks. In 2026, static runbooks will increasingly be replaced by agent-driven operational logic.
AI agents will ingest real-time telemetry, configuration state, historical incidents, and intent models to determine not just what failed, but why. Rather than following a linear checklist, agents will explore multiple hypotheses, validate them against live data, and coordinate corrective actions through orchestrated workflows.
This doesn’t eliminate human oversight – it elevates it. Engineers will move from executing steps to supervising systems, refining intent, and approving high-impact changes. Operations become less about reaction and more about continuous optimization.
Prediction 4: The Skills Divide Will Become Structural
The networking industry has had ample time to adapt. Automation has been mainstream for over a decade. AI-assisted workflows have been viable for three years. Retrieval-augmented generation (RAG) and knowledge grounding have matured over two. MCP has been available for a year.
By 2026, patience will run out.
Organizations will increasingly reduce roles that rely solely on manual execution and institutional knowledge, while expanding roles focused on automation design, orchestration, and agent supervision. This is not about replacing people with AI; it is about replacing non-scalable practices with systems.
Engineers who embrace automation, understand orchestration, and can reason about agent behavior will thrive. Those who resist these shifts will find fewer opportunities. The industry will move forward, with or without them.
Prediction 5: Networking Becomes Intent-First, Not Device-First
As agents and orchestration mature, networking will finally decouple from device-level thinking. Engineers will define outcomes – performance, security posture, resilience – and let agent-driven systems determine the optimal execution across vendors, domains, and layers.
This transition has been promised before, but 2026 will be different. The combination of MCP, agents, A2A protocols, and orchestration provides the missing pieces to make intent-driven networking operationally viable at scale.
Closing Thoughts
By 2026, networking will no longer be measured by how quickly humans can respond, but by how well systems can reason, coordinate, and execute. The organizations that succeed will be those that treat AI agents and orchestration not as add-ons, but as foundational infrastructure for modern network operations.
##
ABOUT THE AUTHOR
John Capobianco is the Head of AI & Developer Relations at Itential, and a technology leader, developer advocate, and builder at the intersection of AI and network automation.
With a career spanning enterprise, government, and cloud networking, John has held roles including Head of Developer Relations at Selector AI, where he focused on AI-driven observability, configuration intelligence, and autonomous network operations, as well as Cisco AI Technical Leader and Senior Network Architect for the Parliament of Canada / House of Commons. He brings deep, hands-on experience applying automation and AI in highly regulated, mission-critical environments. His work centers on helping large organizations adopt AI safely while maintaining reliability, security, and operational trust.
John is a former professor at St. Lawrence College, an author, speaker, and educator. He is the author of Automate Your Network (self-published, 2019) and the Cisco Press pyATS book (2024). He regularly shares insights through talks, workshops, and the Automate Your Network brand, with a focus on practical, production-ready AI, developer empowerment, and the evolution of network engineering in an AI-first world.
