财

Cisco's earnings report sends a new signal: $9 billion in AI revenue, with data center networks taking the baton

LongbridgeAII'm LongbridgeAI, I can summarize articles.

As AI data centers continue to scale up, what's needed beyond just adding more GPUs? Cisco's latest earnings report points to the network connecting compute nodes as the key missing piece. The ability of compute devices to coordinate efficiently is now impacting the overall performance of entire clusters, driving new equipment procurement demand.

On November 12 (local time), Cisco reported its fiscal Q1 2027 results for the period ending October 24, 2026: quarterly revenue reached approximately $18.5 billion (+24% YoY), and the company raised its full-year revenue and non-GAAP EPS guidance. More notably, it increased its full-year AI infrastructure revenue outlook for hyperscale cloud providers. CEO Chuck Robbins described the current industry landscape as being in the early stages of a "multi-year networking supercycle."

Cisco's earnings report sends a new signal: $9 billion in AI revenue, with data center networks taking the baton

Why does networking matter more as GPU clusters grow?

For the past two years, AI infrastructure investment has focused heavily on GPUs, advanced packaging, HBM, and servers. However, in large-scale training clusters, compute nodes frequently exchange data. Insufficient bandwidth, network congestion, or transmission latency can cause GPUs to stall while waiting. Simply increasing the number of GPUs doesn't guarantee a proportional rise in actual compute efficiency.

Switches, high-speed interconnects, and optical connections handle data transfer between nodes. As clusters expand, networking must keep pace; otherwise, newly added compute resources cannot be fully utilized. When customers procure equipment, they must weigh device performance against whether network upgrades can reduce wait times and shorten task completion cycles.

This demand shift is already reflected in Cisco's revenue mix. Network business revenue hit $11.132 billion this quarter (+43% YoY); product revenue grew 32%, and service revenue grew 1%. However, since network business also covers traditional enterprise markets, not all of its revenue or incremental growth should be attributed to AI.

Cisco has accumulated extensive products, customer bases, and operational experience in enterprise networking, and its self-developed Silicon One chips are already deployed in AI networking products. While existing technology helps the company participate in new projects, large clusters demand higher communication efficiency, compatibility, and stability. Whether Cisco's established customer base will translate into new procurement share depends on specific deployment outcomes.

How significant is the $9 billion upside?

Cisco raised its full-year AI infrastructure revenue forecast for hyperscale cloud providers from $7.5 billion to $9 billion for fiscal 2027. In the previous fiscal year, revenue under this metric was approximately $4 billion. If achieved, this target would more than double the prior year's figure. Note that this $9 billion is an annual projection, not yet realized, and applies only to AI infrastructure business with specific customer segments.

For Cisco, hyperscale cloud providers expanding their data centers create procurement needs distinct from traditional enterprise IT refresh cycles. The volume and delivery schedule of individual projects may significantly impact quarterly revenue. Once inside these customers' supply chains, the company must rely on consistent delivery and technological iteration to secure follow-on expansion orders.

Has this growth translated into profitability? GAAP operating profit stood at approximately $5.6 billion this quarter (+66% YoY), with an operating margin of 30.2%; non-GAAP operating margin reached 37.8%. As revenue expanded, GAAP operating expenses decreased 3% YoY, and non-GAAP operating expenses grew 2%, both lagging behind revenue growth, thereby supporting operating profit expansion.

However, margins present another side. GAAP gross margin declined from 65.5% in the prior-year period to 63.7%, while non-GAAP gross margin fell from 68.1% to 65.6%. Faster overall revenue growth and higher aggregate operating margins do not imply every product line is more profitable. Current disclosures are insufficient to isolate AI-specific gross margins; product mix and cost changes require further information for accurate assessment.

Operating cash flow totaled approximately $4.4 billion (+36% YoY), funding procurement, R&D, and daily operations. As large project deliveries increase, the alignment between collections and inventory buildup will determine how much cash remains from revenue growth.

Based on current performance, Cisco raised its full-year revenue guidance to $74.8–$75.8 billion and non-GAAP EPS guidance to $5.47–$5.53.

What hurdles remain after strong demand?

First is customer build-out pacing. AI data centers require coordinated compute, networking, and power infrastructure. Delays in any segment can affect equipment procurement and revenue recognition. While hyperscalers have continuous expansion needs, they adjust budgets and deployment schedules based on project progress, meaning networking vendors' quarterly results may not remain perfectly smooth.

Competition is also intensifying. Arista continues launching high-speed Ethernet products tailored for AI clusters, and NVIDIA is expanding its networking platform. Cisco and NVIDIA maintain product collaborations, so industry dynamics shouldn't be oversimplified as zero-sum competition. Customers ultimately choose solutions based on performance, software adaptability,运维 costs, and delivery capabilities.

How much support does the contract backlog provide? At quarter-end, Cisco's Remaining Performance Obligations (RPO) stood at $46.2 billion (+8% YoY). This metric reflects contracted but unrecognised revenue across all products and services, not exclusively AI orders, and won't convert to revenue in a single quarter. The recognition rhythm matters more for understanding short-term performance than the total amount itself.

Cisco's services business remains another revenue stream. After equipment delivery, customers still need maintenance and upgrades; long-term service relationships help the company continue participating in network updates. However, service revenue growth this quarter significantly lagged behind products. It will take longer to observe whether new equipment procurement drives subsequent service demand.

Conclusion

As AI applications extend from training to broader deployment scenarios, networking products face new demands. If Cisco can leverage its large-cluster delivery experience across more customer projects, traditional networking technologies could unlock significantly larger application spaces.

Sources

1. Cisco Fiscal Q1 2027 Earnings Release (Nov 12, 2026): Current financial data, full-year AI revenue outlook, and guidance.

2. Cisco Fiscal 2026 Full-Year Earnings Release (SEC Filing): Prior fiscal year AI infrastructure revenue and previous expectations.

3. Cisco Technical Briefing: Relationship between AI cluster communication and network performance.

4. Cisco-NVIDIA Product Collaboration Announcement (Mar 2026): AI networking products and partnership details.

5. Arista AI Networking Product Announcement (Oct 2026): Rack-level networking product layout.

Data current as of November 13, 2026. All amounts in USD. Quarterly financials are unaudited; fiscal year follows Cisco's reporting convention; non-GAAP figures use company-adjusted metrics.

The copyright of this article belongs to the original author/organization.

The views expressed herein are solely those of the author and do not reflect the stance of the platform. The content is intended for investment reference purposes only and shall not be considered as investment advice. Please contact us if you have any questions or suggestions regarding the content services provided by the platform.