DIVE BRIEF:
- AI adoption is quickly becoming a campus and branch networking problem, according to a study published by Cisco and market research firm Foundry. The report, based on a survey of nearly 3,500 CIOs and IT leaders, revealed that 73% of organizations already face, or expect to face, capacity limitations in their campus and branch networks within the next 24 months.
- Pressure is mounting. Respondents said AI-driven traffic tied to campus and branch networks rose 34% in the last 12 months. They expect it to increase 96% over the next year and 209% over the next three years, as generative, agentic and physical AI workloads surge.
- AI readiness is moving beyond GPUs, cloud platforms and data center buildouts to the core infrastructure, creating modernization and managed service opportunities for partners. The study points to a need to build resilience and flexibility into busy workplace networks, which weren’t built to handle the scale and unpredictability of AI traffic.
DIVE INSIGHT:
The challenge facing networks and the partners that upgrade and maintain them isn’t just more traffic — it’s the composition and nature of the workloads.
Two-thirds of respondents reported increases in east-west traffic tied to AI workloads, according to the study. Only marginally fewer noted increased real-time or latency-sensitive traffic and 61% had already seen growth in continuous automated system traffic.
“Usually, networks are designed for consistent traffic, like SaaS and CRM traffic, and there aren’t a lot of unpredictable traffic patterns,” the head of AI strategy for global IT and network engineering operations at a large U.S. technology company told Cisco and Foundry in the report. “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?”
The question opens doors for channel partners. Customers need more than bandwidth. They need help understanding where AI traffic is showing up, how it behaves, which workflows are latency-sensitive and where security controls are weakest. The challenge calls for a range of services, including network assessments, campus and branch modernization, Wi-Fi upgrades, segmentation, observability, automation and managed security services.
Over three-quarters of respondents acknowledged their campus and branch environments require upgrades to support current and future AI-driven workloads. More than 9 in 10 said they were accelerating modernization initiatives in response to AI-driven demand.
Even the most aggressive AI adopters aren’t ready. Only 30% of organizations with broad enterprise-wide deployment of generative AI were confident they could support projected AI growth across their networks, according to the report.
Security complexity remained a persistent headache associated with network demand. More than three-quarters of respondents expect risks to increase as AI expands beyond generative use cases.
“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,” a U.S.-based retail executive said in the report.
The channel role extends beyond replacing legacy hardware. Partners have an opportunity to step in as AI-enablement advisers, building customer networks with visibility, security and performance to support AI as it moves from pilot projects into everyday operations.
The business case is already becoming clear. Cisco and Foundry found that 75% of IT leaders recognize that modernization delays can lead to higher long-term costs due to reactive upgrades and remediation.
Clients may still be focused on models, copilots and automation, but the partner opportunity sits underneath the application layer, across the branch, campus and wireless networks that will determine whether AI can scale.
Network modernization is no longer simply an infrastructure initiative. It is becoming a prerequisite for operating and competing effectively in an AI-powered economy,” the report concluded.
