While enterprise IT leaders have spent the past two years focusing AI infrastructure discussions on GPUs, cloud platforms, and data centers, new Cisco research suggests that enterprise networks may not be ready for the next phase of AI adoption. The study, conducted in partnership with Foundry, surveyed 3,472 IT and networking leaders across 15 countries. It found that artificial intelligence is already reshaping traffic patterns in campus and branch environments, exposing critical gaps in capacity, security, and visibility that many organizations are ill-prepared to address.

"We have entered a networking supercycle, because the network is so central to all the AI infrastructure the world is building now," said Jeetu Patel, Cisco president and chief product officer, in a statement. This sentiment underscores a pivotal shift: the conversation around AI readiness must expand beyond data centers and cloud environments to include the networks that connect employees, applications, and devices. The issue will become more acute as organizations move from generative AI pilots to widespread deployment of AI agents that communicate continuously with other systems and applications.

Key Findings from the Cisco Survey

The survey revealed several startling statistics. Organizations reported a 34% increase in AI-related campus and branch network traffic over the past 12 months. This growth is projected to accelerate dramatically, with traffic expected to climb by 209% over the next three years. Companies that are aggressively deploying AI anticipate their total network traffic will triple during that period. Furthermore, 73% of respondents already face, or expect to face, campus and branch network capacity constraints within the next two years.

These findings highlight a fundamental challenge: traditional network architectures, often designed for predictable traffic patterns like SaaS and CRM applications, are ill-equipped to handle the dynamic and often unpredictable flows generated by AI workloads. The survey also found that 67% of organizations say AI workloads are increasing east-west traffic between internal systems and applications. This lateral traffic, common in distributed AI processing, puts additional strain on network segments that were not originally dimensioned for such demands.

Implications for Networking Teams

Changing traffic patterns inside enterprise environments are causing additional pressure for enterprise network teams. One participant in the research, the head of AI strategy for global IT and network engineering operations at a large U.S. technology company, explained the predicament: "Usually, networks are designed for consistent traffic, like SaaS and CRM traffic, and there aren't a lot of unpredictable traffic patterns. Suddenly, three AI agents are trying to talk to each other and solve a problem. That is going to be a big thing … how do we support increased east-west traffic?"

Cisco defined aggressive AI adopters as organizations with broad generative AI deployments across the enterprise. However, only 30% of those organizations said they are fully prepared to support projected AI growth across their networks. As a result, 93% of IT decision makers said they are accelerating network modernization efforts. This indicates a widespread recognition that the network must evolve to remain a viable foundation for AI-driven operations.

Security and Observability Challenges

The report also highlighted an observability challenge that could complicate future deployments. As employees and business units increasingly experiment with AI tools, IT organizations often lack visibility into what is actually running on their networks. The same AI strategy executive noted, "Right now, we don't even know what the AI-driven demand is. Observability is a huge gap. There is experimentation going on all over the place, and there is no way for us to really identify if somebody is deploying some kind of service on our network, whether it is a genAI solution or an agentic solution."

Security is also emerging as a major barrier to AI expansion. The survey found that 80% of respondents said AI has expanded their attack surface, and 61% said they are delaying additional AI deployments until they gain more confidence in their security posture. The vice president of infrastructure, network, and end-user services at a U.S. retail enterprise interviewed for the report commented, "The issue from a security standpoint is that it's hard to create the guardrails for every possible AI tool that your organization must use."

These security concerns are compounded by the rapid proliferation of AI agents. The survey found that 85% of respondents expect moderate or significant growth in AI agent deployments over the next two years. Each agent represents a potential endpoint that must be secured and monitored, further straining existing security frameworks.

AI Readiness Beyond Data Centers

The AI readiness conversation has often centered on data centers, but AI applications operate where employees work, devices connect, and business processes run. That means campus and branch environments may become just as important to AI success as the infrastructure supporting AI models. Organizations that neglect their edge networks risk undermining the very benefits AI is supposed to deliver—increased productivity, faster decision-making, and enhanced customer experiences.

To address these challenges, enterprises are investing in network modernization initiatives. Many are exploring technologies like SD-WAN, AIOps, and network segmentation to improve performance, visibility, and security. Some are also adopting zero-trust architectures to better control access to AI resources and data. However, the pace of change must accelerate if organizations are to keep up with the exponential growth of AI traffic.

The Cisco research shows that AI infrastructure planning can no longer focus only on back-end systems if enterprises expect to scale AI deployments over the next several years. As Patel said in the statement: "Eventually there will be only two kinds of companies: those that are AI companies, and those that are irrelevant." The network, it seems, will be a critical determinant of which category a company falls into.

For more context, the findings align with broader industry trends. Networking vendors are increasingly embedding AI capabilities into their products, from predictive analytics to automated troubleshooting. Meanwhile, enterprise leaders are recognizing that the network is not just a plumbing layer but a strategic asset that must evolve in tandem with AI adoption. The challenge is that many organizations have underinvested in campus and branch networks for years, focusing instead on data center and cloud connectivity. This legacy approach is no longer sustainable.

As AI agents become more sophisticated and numerous, they will demand not only more bandwidth but also lower latency and higher reliability. Network teams must prepare for a world where AI-to-AI communication becomes the norm, requiring real-time prioritization and dynamic traffic engineering. This will likely drive adoption of advanced technologies like time-sensitive networking (TSN) and intent-based networking (IBN).

In summary, the Cisco survey serves as a wake-up call for enterprises that have been slow to modernize their campus and branch networks. With AI traffic set to more than triple in the next three years, and with the majority of organizations already facing capacity constraints, the time to act is now. Those that invest in network visibility, security, and scalability will be better positioned to harness the full potential of AI. Those that delay may find their networks become a bottleneck rather than an enabler.

Source: Network World News