Wall Street just gave Cisco a massive thumbs up, and honestly, it’s about time. For the last couple of years, the networking giant looked like it was stuck in the slow lane while Nvidia and Microsoft hogged all the AI glory. That changed in a heartbeat this week. Shares of Cisco popped 17% after a stellar earnings report that proved one thing: the AI infrastructure boom is no longer just a theory for the "old guard" of tech.
It’s a weirdly bittersweet moment, though. While investors are popping champagne, nearly 4,000 employees are packing their desks. This isn't just about a bad quarter; it’s a cold, calculated shift in how the company spends its money. Cisco is betting the house on AI, and they’re willing to trim the fat to get there.
The billion dollar AI surge
Let’s look at the numbers because they’re kinda ridiculous. Cisco didn't just beat estimates; they blew them out of the water. The company reported roughly $2.1 billion in AI-related product orders in the latest quarter alone. To put that in perspective, that’s up from $1.3 billion just the quarter before.
Management is now signaling that AI hardware bookings could blast past $9 billion for the full fiscal year. That’s a massive jump from their previous $5 billion target. It turns out that when hyperscalers like Amazon, Google, and Meta build out their massive data centers, they need more than just GPUs. They need the networking "pipes" to connect those chips, and that’s where Cisco’s Silicon One and Nexus 9000 switches come in.
The human cost of a pivot
You can’t talk about this stock rally without mentioning the layoffs. Cisco is cutting around 7% of its global workforce, which works out to about 4,000 to 5,900 people depending on the final tally. It’s the second major round of cuts this year.
Why do this when you're making billions? It’s basically about "reallocating," as CFO Scott Herren put it. They aren't just trying to save cash; they’re trying to move hundreds of millions of dollars from legacy departments—think old-school campus networking—into the fast-growing buckets of AI, cloud, and cybersecurity.
If you’re working in a department that isn't helping a server talk to a GPU at lightning speed, your seat is getting a lot hotter. Cisco is effectively becoming a software-first company that happens to sell high-end hardware on the side.
The Splunk factor
The $28 billion acquisition of Splunk is finally starting to show its teeth. Security revenue was up 81% this quarter, largely thanks to Splunk’s contribution of about $960 million. By merging networking, security, and observability into one unit under Jeetu Patel, Cisco is trying to solve the biggest headache for IT departments: "How do I secure all this new AI data without it slowing down my network?"
What people get wrong about Cisco’s AI play
Most casual observers thought Cisco missed the boat because they don't make the actual AI chips. That’s a huge misconception.
AI isn't just a compute problem; it’s a connectivity problem. When you have thousands of GPUs working on a single training model, the bottleneck often isn't the processing power—it’s the speed at which data moves between those processors. Cisco’s Ethernet-based fabrics are now competing directly with InfiniBand (a rival tech often favored by Nvidia). If Ethernet wins out for its flexibility and lower cost, Cisco becomes the toll booth for the entire AI industry.
Is this sustainable for your portfolio?
If you're looking at that 17% jump and wondering if you've missed the trade, consider the guidance. Cisco raised its full-year revenue outlook to a range of $62.8 billion to $63 billion. They’re also projecting that AI orders will stay hot through 2026.
The "sick man" of Silicon Valley label is officially dead. But keep an eye on those gross margins. As they ramp up AI hardware, margins can get squeezed compared to their high-margin software subscriptions. The goal for CEO Chuck Robbins is to prove that they can sell the hardware now and attach enough recurring security software to keep the profits growing long-term.
If you’re managing an enterprise network or an investment portfolio, the takeaway is simple. Don't bet against the infrastructure. While everyone fights over which AI app will win, the guys building the roads—like Cisco—are the ones getting paid regardless of who wins the app war. Check your exposure to legacy networking; if it’s not AI-optimized, it’s a liability.
