Cisco is positioning itself to become a primary supplier of backend infrastructure and networking for enterprise AI datacenters.

Cisco has articulated a clear strategy to capitalize on its datacenter infrastructure strengths for the AI market. In the past quarter, the company made $300 million from AI-specific products and says it's on pace to hit $1 billion in AI revenue in fiscal 2025—a notable milestone, though modest relative to Cisco's $50+ billion in annual revenue. Jeetu Patel, named the company's chief product officer last summer, has championed a "platform advantage" strategy where Cisco's different offerings work together cross-functionally. Drawing an analogy to personal electronics, Patel noted his preference for Apple products—iPhone, MacBook, iPad, and Apple Watch—which work seamlessly together, making customers more likely to choose additional products from the same maker. This same logic underpins Cisco's approach to enterprise AI infrastructure.

Cisco's existing strengths position it well for this platform strategy. The company maintains an established presence in networking and cybersecurity, has expanded its data management capabilities through acquisitions including Splunk, and is increasingly ambitious about datacenter compute. Patel stated clearly: "We are going to double down on the compute business" and "We are in the compute business unapologetically." However, the margins on compute are substantially lower than Cisco's other businesses, so the company's strategy appears focused on picking specific niches rather than competing at scale with Dell, Lenovo, and HPE. The "better together" philosophy works for customers who value this integrated approach within complex enterprise AI deployments.

Two major infrastructure products announced by Cisco exemplify this strategy. The Nexus 9000, an 800-gigabit switch rooted in Cisco's core networking competence, is already being used by hyperscalers. Alongside it, Cisco offers a complementary server with eight Nvidia GPUs for AI training, built on the Unified Computing System approach that combines compute, networking, and storage. These products support Patel's assertion that "We are now in the AI infrastructure business."

The company has notched at least one big design win for hyperscaler AI and sees continued momentum in back-end networking for hyperscaler LLM training clusters. Yet Patel points out that something like $200 billion has been spent industrywide on AI training so far to yield something like $5 billion to $10 billion in revenue. As enterprises transition from experimentation to mass deployment of AI applications, Cisco is positioning itself to capture meaningful IT spending in inference and enterprise AI deployment—areas expected to ramp up seriously in 2025, likely in the second half of the year.