Cisco is repositioning itself as a core AI infrastructure supplier, reporting strong hyperscaler orders and a broadening market beyond GPUs to networking, security, and observability. The company raised its FY26 hyperscaler AI infrastructure order expectation to ~$9B and highlighted Silicon One as a critical differentiator.
Cisco is saying the AI “marketplace” is becoming a networking-led infrastructure cycle, not just a GPU/server cycle. The company’s core message: AI workloads create huge east-west traffic, security exposure, observability needs, and campus/data-center refresh demand — all of which should pull through Cisco switching, Silicon One, optics, security, and Splunk. **What Cisco is saying** **Theme** **Cisco’s message** **Evidence** Hyperscaler AI demand is real and accelerating Cisco raised its FY26 hyperscaler AI infrastructure order expectation to \~$9B, up from >$5B earlier; Q3 hyperscaler AI infrastructure orders were $1.9B, vs $600M a year earlier. AI demand is broadening beyond hyperscalers Cisco says it took \~$300M of Q3 AI infrastructure orders from neocloud, sovereign, and enterprise customers, with a \~$3B pipeline across those customers. Enterprise AI is still early Management calls enterprise AI infrastructure a ramping opportunity, with customers preparing for inferencing and agentic applications, not just model training. Cisco sees a multi-year refresh cycle Cisco argues legacy infrastructure was not built for AI’s performance, speed, and security demands, creating a “once-in-a-generation” transition. Silicon One is strategically critical Chuck Robbins said Cisco’s $9B AI target would “probably be close to zero” without its own silicon — meaning Silicon One is not a side story; it is the hyperscaler entry ticket. AI is also a security + observability opportunity Cisco is tying AI to AI Defense, zero-trust for AI agents, agentic SOC, Splunk observability, and broader secure networking. **Key management commentary** ***“With this AI revolution, everything is dependent upon a network.”*** ***— Chuck Robbins, CSCO, J.P. Morgan TMT Conference 2026*** ***“Legacy infrastructure was not designed for the performance, speed, and security needs of AI.”*** ***— Chuck Robbins, CSCO, Q2 FY2026 earnings call*** ***“If we didn’t have our own silicon, the $9 billion that we announced would probably be close to zero.”*** ***— Chuck Robbins, CSCO, J.P. Morgan TMT Conference 2026*** **My read** Cisco is trying to reposition itself as one of the picks-and-shovels winners of AI infrastructure. The most important thing is that Cisco is no longer talking about AI in vague terms. It is putting hard numbers around orders: $5.3B year-to-date hyperscaler AI infrastructure orders by Q3 FY26, a new \~$9B FY26 order target, and \~$4B expected FY26 AI infrastructure revenue from hyperscalers. That is a meaningful shift from “AI narrative” to measurable backlog/order conversion. CSCO Q3 FY2026 earnings call The second important point: the market is still concentrated but beginning to broaden. Hyperscalers remain the biggest driver, but Cisco is now explicitly calling out neocloud, sovereign cloud, enterprise, telco, and campus refresh demand. That matters because the bull case needs AI to move beyond a handful of cloud customers into mainstream enterprise infrastructure budgets. The strongest part of Cisco’s positioning is Silicon One + optics + networking systems. Management is effectively saying: AI clusters and inter-data-center AI traffic require specialized networking, and Cisco can win because it controls silicon, systems, and optics. The weaker part is that enterprise AI revenue is still early — the enterprise/neocloud/sovereign pipeline is promising, but not yet as proven as hyperscaler orders. **Bottom line** Cisco’s message is: AI turns the network into critical infrastructure again. If that proves right, CSCO is not just a legacy networking vendor participating at the edges — it becomes a core AI infrastructure supplier. But the stock’s AI thesis still depends on two things: continued hyperscaler order conversion and proof that enterprise AI infrastructure spending broadens beyond pilot projects into a real multi-year refresh cycle. Sources: • CSCO Q3 FY2026 Earnings Call • CSCO Q1 FY2026 Earnings Transcript • CSCO J.P. Morgan TMT Conference - 2026-05-18 • CSCO Q2 FY2026 Earnings Call https://knowledge.dotadda.io
Cisco and OpenAI collaborated to transform Codex from a code completion tool into an enterprise-grade AI engineering agent capable of autonomous workflows at scale, achieving significant productivity gains including 20% build time reduction and 10-15× improvement in defect resolution throughput.
Cisco is cutting nearly 4,000 jobs (5% of workforce) to reallocate spending toward AI and cybersecurity, despite reporting record quarterly revenue and better-than-expected profit.
Coherent breaks ground on an expanded facility in Texas, backed by a $50 million CHIPS Act grant, to scale production of optical components and compound semiconductors essential for AI infrastructure. NVIDIA CEO Jensen Huang highlighted the critical role of optics in enabling large-scale AI systems.
AI infrastructure spending continues to accelerate, particularly in data centers and advanced chip production, with semiconductor testing equipment companies like Teradyne positioned as potential overlooked winners.
This report from Multiples.vc provides public AI valuation multiples as of June 2026, covering hyperscalers, semiconductor supply chain, neoclouds, and other segments with median forward revenue multiples and growth rates.