Network operations teams are drowning in complexity, and many now see autonomous AI as the lifeline. According to a new survey of 1,000 IT and network operations leaders, 80% are comfortable giving AI a high or fully autonomous role in network operations. The findings come as enterprise networks become too complex for humans to manage alone, driven by rapid change, cross-domain dependencies, and the explosive growth of AI traffic itself.
The survey, conducted by a major networking vendor and a research and advisory firm, paints a picture of an industry ready to cede some control to AI agents. Three-quarters of respondents already use AI in some fashion for network operations. More than half (51%) use agentic AI tools in production to take corrective action in real time, rather than simply taking advice. Eighty-four percent expect to reach a fully AI-led operating model within twelve months.
The comfort level varies. While 80% are open to high or fully autonomous AI, 56% will do so only with human approval of those actions. Another 24% are comfortable with AI taking network actions with no human oversight. The vast majority (82%) are comfortable allowing AI to make some production network changes on its own for certain categories of tasks.
This shift is being called a move from AIOps to agent-powered operations, or AgenticOps. Traditional AIOps tools have helped surface anomalies and provide recommendations, but they often require too much human interpretation. Agentic AI goes further, using autonomous agents to diagnose issues, decide on a course of action, and execute fixes within predefined workflows.
Unlike generative AI that responds to prompts, agentic AI can plan, act, and adapt. In network operations, that means monitoring telemetry, correlating events, identifying root cause, and executing remediation. It can operate within guardrails, but it can also learn from outcomes to improve over time. The promise is not just faster response but also the ability to handle the sheer volume of events that human teams cannot.
Industry analysts say the move is inevitable. One analyst noted that networking professionals will embrace agentic operations because the alternative is unsustainable. Trust will take time, similar to the curve for autonomous vehicles, which get into far fewer accidents than human drivers. Agentic AI tools will make mistakes, but far fewer than people, and they will free staff to focus on higher-value work.
Another analyst attributed the shift to a combination of increasing network complexity and a decreasing number of humans with the right skills to solve networking problems, especially cross-domain issues. As networks span cloud, security, applications, and endpoints, traditional siloed expertise is no longer enough. The skills gap is not just about headcount; it is about the breadth of knowledge required to troubleshoot modern networks.
AgenticOps Requires Effective Guardrails
The push toward autonomy comes amid broader calls to slow AI adoption, partly in response to high-profile incidents where autonomous agents ran amok. Yet the survey shows hundreds of network professionals are willing to cede at least some control, provided strict guardrails are in place. Nearly every respondent (99%) said they would not trust AI to act without such safeguards.
The guardrails respondents demand include:
- Explainable AI actions
- Human approval for actions
- Policy-based operational limits
- Emergency override mechanisms
- Role-based access control
- Immutable audit trails
These guardrails build trust by letting users see the reasoning that leads an agent to a conclusion. While AI agents can draw conclusions from the intelligence they examine, any actions they take map to predefined workflows based on the network team's standard operating procedures for different situations. That means the agent is not improvising; it is executing a playbook, albeit at machine speed.
Another key constraint is that agentic agents do not talk to other agents. That helps because when they are constrained to only figuring things out themselves, they are less dangerous. Agents are also built around specific skills or knowledge and a defined scope of responsibility. When done on a product basis, this can be controlled effectively. The combination of scoped agents, no agent-to-agent chatter, and human-defined workflows reduces the risk of runaway automation.
When these constraints are combined, they give enterprise network operations teams the confidence to leverage agentic AI to manage increasing complexity. The survey suggests that confidence is already translating into production use, especially for routine tasks where the same corrective action has been performed many times before.
Driving the Need: Complexity, Rapid Change, and AI Itself
Complexity is indeed increasing. Fifty-nine percent of survey respondents report making changes to their production network environments at least daily. Half of those organizations make multiple changes per day, and for a meaningful share, change happens multiple times per hour. As a result, 57% say their change processes cannot keep up. Manual change management, with its approvals and documentation, simply cannot match the velocity of modern network operations.
Ninety-two percent say performance issues tend to cross multiple domains, including cloud, security, applications, and endpoints. Similarly, 95% say their existing, non-agentic tools fall short in significant ways, mostly by requiring too much human interpretation and lacking cross-domain visibility. When a problem spans cloud infrastructure, a security policy, and an application dependency, no single tool provides the full picture. Engineers must piece together data from multiple sources, often under time pressure.
The generative AI boom has also increased network complexity. Two-thirds of respondents say so. Traffic analysis of direct-to-AI traffic shows average daily AI traffic is on a trajectory to double every six months. This acceleration is likely fueled by the growing complexity of AI tasks, which demand more data exchange than a simple query. AI models are not just answering questions; they are processing images, audio, video, and large datasets, all of which put new demands on network capacity and latency.
The result is an alert deluge. The average organization generates around 4,100 monitoring alerts and events per day, with 51% being network-related. A typical practitioner can review, investigate, and resolve about 21 network alerts a day, meaning it would take a team of about 100 specialists to handle that daily volume. That math is impossible for most organizations, which explains why automation is no longer optional.
Since few organizations have that much staff, nearly half of alerts (46%) are closed without investigation. Alert fatigue is a meaningful source of employee dissatisfaction for 65% of respondents, while 67% say alert volumes prevent teams from doing other critical work. One analyst said those numbers ring true based on research around security alerts, which found well under 50% of alerts are investigated. The pattern is clear: too many alerts, too little time, and too much risk.
Too Many Tools, Too Much Time to Resolve Issues
Organizations rely on an average of 10 tools to try to maintain end-to-end visibility, but they tend to be siloed, making it difficult to diagnose problems that cross domains. That is reflected in the time it takes to resolve issues. The mean time to resolve a network incident is 88 hours, while the median is 12.5 hours. The mean is skewed by some organizations that take a week or longer, reflecting the degree of complexity. The median suggests that many incidents are resolved relatively quickly, but the long tail is costly.
One analyst sees a real opportunity to do better by having AI help with analysis and automate the root cause process. One example is when an organization tends to take the same corrective action every time a situation occurs. If a fix has been done 14 times, it can be automated, but the system should still notify the team that it happened. This kind of automation preserves human awareness while eliminating repetitive toil.
Network professionals may also take a page from their security counterparts. With a similar problem of too many issues to tend to, security pros are increasingly automating the response, even if that means shutting down a resource. The thinking is that potential losses are greater than the potential impact to the business. Network teams are moving toward being more comfortable with that approach, especially for well-understood failure modes where the remedy is clear.
Automation can also help with cross-domain issues. If an agent can correlate data from network, security, and application tools, it can identify root cause faster than a human who must switch between consoles. The agent can then execute a workflow that might involve changing a routing policy, adjusting a firewall rule, or restarting a service. With proper guardrails, these actions can be taken safely and audited.
The Solution: Another Single Pane of Glass
The networking vendor behind the survey is proposing a new cloud control platform as a solution. It is intended to provide a unified view and management plane for networking, security, compute, observability, and collaboration solutions. The platform also applies agentic AI to diagnose and resolve issues, including those that cross domains. It is, yet again, the proverbial single pane of glass, this time with an AI twist.
It may seem ironic that a vendor that sells networking gear that has become too complex to manage is now also selling the solution intended to address that complexity. One analyst acknowledged that network vendors have been complicit in creating complexity but said it is good to see them simplify things now. The network is being used in many more ways than ever before, supporting orders of magnitude more devices and connecting to nearly everything.
Another analyst agreed, noting that watching networks evolve and attempts to automate operations has been very difficult. But with AI, the industry may finally be getting there. The survey suggests that network professionals are not just ready for autonomous AI; they are counting on it to keep their networks running as complexity continues to accelerate. The question is no longer whether AI will play a larger role in network operations, but how quickly organizations can implement the guardrails that make autonomy safe and trustworthy.
As networks become more critical to every aspect of business, the ability to manage them at machine speed will become a competitive advantage. The survey shows that the workforce is ready to embrace that future, provided the technology respects human oversight, explains its actions, and operates within clear boundaries. With those conditions met, agentic AI could transform network operations from a reactive cost center into a proactive, self-healing utility.
Source:Network World News