In a surprising revelation during Cisco’s recent earnings conference call, CEO Chuck Robbins shared an impressive milestone: the company had already managed to secure sales exceeding half a billion dollars in AI equipment. This unexpected announcement showcased Cisco’s strategic efforts to establish a strong presence in the burgeoning AI market, catching analysts and industry insiders off guard.
Cisco’s innovative approach involved targeting cloud computing giants, often referred to as hyperscalers in the tech world, and offering them cutting-edge AI solutions. Notably, these cloud behemoths were drawn to Cisco due to its expertise in Ethernet, a well-established networking technology synonymous with the brand. The pivot towards Ethernet for AI applications represents a significant shift from the prevalent use of Infiniband, a high-speed networking technology usually favored by data centers. This maneuver prompted Cisco’s unexpected collaboration with industry players like AMD, Arista, Broadcom, HPE, Intel, Meta, and Microsoft, collectively forming the Ultra Ethernet Consortium. The consortium’s mission is to enhance Ethernet performance, aiming to double the speeds achieved by the swiftest 400-gigabit connections currently available. Simultaneously, the consortium focuses on adapting Ethernet technology to better accommodate the demands of AI-driven processes.
During the earnings call, Robbins stated, “To date, we have taken orders for over half a billion dollars for AI Ethernet fabrics. We are also piloting 800 gig capabilities for AI training fabrics.” While the figure might appear relatively small in the context of Cisco’s financial standing, marked by an impressive $57 billion in sales for the fiscal year 2023 and an earnings per share of $3.07, Robbins highlighted the substantial growth potential of this emerging AI cloud market.
Drawing parallels to the past, Robbins candidly acknowledged Cisco’s previous misstep in capitalizing on the original cloud boom. He emphasized, “This would probably be three to four times the opportunity size of the original cloud build out. And, you know, unfortunately for us, as it’s been well documented, we missed the original cloud build out.” He alluded to the fact that major players like Google and Facebook had opted to develop their data center networking technology or partner with specialized companies like Arista Networks during the initial cloud expansion.
Robbins’s optimism about Cisco’s future prospects in the AI realm was palpable. He confidently asserted, “But I can say with every bit of confidence right now that as we go through this AI transition to Ethernet, we are super well positioned.” He highlighted Cisco’s current involvement in 21 use cases across the top six cloud providers and expressed confidence in sustaining this momentum in the coming years.
While industry analysts like Meta Marshall from Morgan Stanley are intrigued by Cisco’s strategic maneuvering, a degree of caution lingers. Marshall noted in a post-earnings research note that the investment community is closely watching companies that position themselves for the AI infrastructure expansion. She acknowledged the relatively modest revenue from the $500 million in sales thus far but underscored its significance in fostering deeper connections within the hyperscaler customer base. Despite acknowledging that Cisco might have a smaller market share within this space compared to competitors, Marshall recognized the positive trajectory of increased penetration.
As Cisco navigates the evolving landscape of AI and networking, Robbins’s ambitious vision of reaping the rewards from this burgeoning market seems to be driving the company’s determination to seize the opportunities this time around. With a valuation of approximately $225 billion, a quarter of industry titan Nvidia’s $1 trillion valuation, Cisco is positioning itself to stake a claim in the expanding AI domain and secure a more substantial slice of the lucrative market pie.
