The News
Cisco released The Impact of Agentic AI on Network Operations, an independent study conducted by Omdia across 1,000 IT and network operations leaders at organizations with 500 or more employees. The report finds that 51% of organizations already run agentic AI systems that take action, not just provide recommendations, in production network environments today, and 84% expect to reach an AI-led operating model within 12 months. Critically, 80% of IT and NetOps leaders are comfortable granting AI a high or fully autonomous role in network operations, with 24% comfortable with fully autonomous operation requiring no human oversight at all.
Analyst Take
The autonomy gap is closing faster than most vendors expected
The conventional wisdom inside the enterprise networking industry has been that AI in NetOps would spend years in an advisory lane, flagging anomalies and generating recommendations for humans to approve. Cisco’s data punctures that assumption decisively. When 82% of respondents say they are already comfortable allowing AI to make at least some production network changes without prior human approval, the question is no longer whether agentic NetOps is coming. It’s whether the tooling, governance, and vendor ecosystems can keep pace with the appetite organizations already have for it.
This matters for ITDMs because the operational pressure driving that appetite is real and quantifiable. The average organization in Cisco’s study generates approximately 4,100 monitoring alerts and events per day, uses nearly 10 tools to maintain end-to-end visibility, and sees 92% of performance issues span multiple domains. That is not a tooling problem that human escalation chains can solve at acceptable cost. The economics of autonomous action are becoming compellingly obvious to the people writing the budgets.
Trust is the architecture, not the afterthought
The more interesting finding in the Cisco data is not the comfort with autonomy. It’s the conditions attached to it. Ninety-nine percent of respondents require one or more guardrails before trusting AI to act, and 69% specifically require detailed explainability for agent-driven actions. That combination tells a precise story: organizations are not handing over the keys out of blind faith. They are defining a governance architecture first and then expanding the autonomy envelope within it.
For developers building on top of agentic infrastructure, this is a design constraint with teeth. Explainability cannot be a logging afterthought bolted on post-deployment. It needs to be native to how agents reason, surface decisions, and document their actions. The 86% of respondents who say a single integrated platform is the most effective path forward reinforces this point. Fragmented tooling is not just an operational headache; it is structurally incompatible with the audit trails and unified visibility that trust in autonomous systems requires.
What the government sector data tells us about the broader readiness gap
The dynamics Cisco is documenting in enterprise NetOps reflect a broader pattern visible across the public sector as well. ECI Research’s GovTech Survey Results found that 31.8% of respondents identified “FedRAMP/compliance approval friction for AI vendors” as the single largest blocker preventing widespread AI adoption in developer workflows, a finding that maps directly onto the guardrail and governance requirements Cisco’s respondents are articulating. Regulated and security-conscious organizations are not opposed to agentic AI. They are waiting for the compliance scaffolding to catch up with the capability.
That scaffolding gap also shows up in how organizations think about the tools themselves. According to ECI Research, 56.0% of respondents reported that “procurement or contractual requirements frequently force engineering teams to use suboptimal developer tools” because approved vendor lists lack modern developer platforms. Cisco’s push toward an integrated, agent-powered NetOps platform will run directly into this friction in any regulated vertical. The 84% of organizations expecting an AI-led operating model within 12 months is an ambition statement. The procurement and compliance infrastructure in many sectors will stretch that timeline considerably.
The competitive read: Cisco is staking out the platform position early
Cisco’s framing of “AgenticOps” as a defined operating model, with Cisco positioned as the integrated platform that 86% of respondents prefer, is a deliberate market-positioning move. The company is establishing conceptual ownership of a category before the competitive field fully consolidates. That puts pressure on rivals across the AIOps and network management space, from legacy players like IBM and HPE Aruba to cloud-native entrants building on top of hyperscaler observability stacks. ECI Research’s GovTech Survey Results found that only 0.5% of respondents selected “Platform breadth and AI capabilities” as the factor carrying the greatest weight in their final technical selection process, with cost and developer velocity dominating instead. Cisco’s integrated platform argument will resonate most in enterprise accounts where complexity already justifies platform consolidation. In cost-pressured or procurement-constrained environments, the pitch will need sharper economic framing to land.
Looking Ahead
The 12-month window Cisco’s data points to is aggressive, and it is likely that the actual timeline for most organizations achieving a genuinely AI-led NetOps model will extend into 2027 and beyond, particularly in regulated industries where compliance certification cycles add months to any new AI deployment. What will accelerate in the near term is the governance layer. Expect a rapid proliferation of AI agent policy frameworks, explainability requirements baked into vendor contracts, and audit standards that treat autonomous network actions with the same scrutiny as code deployments. The organizations that build that governance infrastructure now will have a meaningful head start when the autonomy dial turns further.
For Cisco specifically, the next test is whether the company can translate compelling survey data into a coherent product narrative that connects its networking hardware, observability software, and AI agent capabilities into a single purchasable outcome. The integrated platform preference documented in this research is an open door. Closing it will require more than research marketing; it will require demonstrable interoperability across Cisco’s own portfolio, which has historically been a challenge. Watch the company’s enterprise customer wins in the next two to three quarters as the real indicator of whether the AgenticOps framing is converting to revenue.
Stay Ahead of Application Development Trends
Get weekly analyst insights, research notes, event coverage, and AppDevANGLE updates delivered directly to your inbox.
Subscribe for Weekly Insights
Join technology leaders, practitioners, and GTM teams following the trends shaping modern software delivery.
Looking for deeper research access?
Explore ECI Research reports, survey insights, and market analysis through the ECI Research Portal.
