I’m writing this from Cisco Live in Las Vegas, where Cisco is making a strong case that it wants to become the infrastructure operating system for AI. There is plenty to discuss in that strategy, and I suspect I’ll be writing about it across several Techstrong properties over the coming days.

What caught my attention this week, however, wasn’t Cisco.

It was NVIDIA.

Not because NVIDIA was sponsoring the event. Every major technology conference has sponsors. Not because NVIDIA had a booth on the show floor. That’s expected. What stood out was how deeply NVIDIA was woven into the conference itself. Joint sessions, executive appearances, product announcements, customer stories and strategic initiatives all seemed to feature NVIDIA somewhere in the discussion.

As I sat through sessions and walked the halls, I realized Cisco Live wasn’t unique.

Over the last year I’ve attended events focused on cybersecurity, DevOps, platform engineering, cloud native technologies, enterprise infrastructure and AI. Different communities. Different audiences. Different business priorities. Yet NVIDIA seems to appear at all of them, often occupying a role that extends far beyond traditional sponsorship.

At RSA, vendors talk about securing AI environments built on NVIDIA infrastructure. At KubeCon, conversations around AI workloads quickly intersect with NVIDIA. At PlatformCon, discussions about developer platforms increasingly involve AI deployment requirements that ultimately connect back to NVIDIA-powered infrastructure. Earlier this year at SUSECON, NVIDIA partnerships surfaced repeatedly in discussions around AI-ready enterprise environments. The same pattern appears at Dell Technologies World, Red Hat Summit and a growing number of industry events that historically had little connection to semiconductor companies.

The obvious explanation is that NVIDIA has become the dominant company in AI infrastructure. Companies naturally want to be associated with market leaders. Customers want reassurance that their technology choices align with where the market is heading. Conference organizers know attendees want to hear from companies shaping the industry’s future.

That’s true, but I don’t think it fully explains what we’re seeing.

What interests me is the way NVIDIA appears to be leveraging its position.

Most technology companies that achieve market leadership eventually become somewhat passive. Their ecosystems carry the message. Their customers become advocates. Their partners handle much of the evangelism. Success allows them to spend less time proving themselves.

NVIDIA seems to be doing the opposite.

The company continues showing up. It continues investing in conference presence. It continues placing executives, engineers and technical experts in front of audiences that are not necessarily attending events specifically to hear about NVIDIA.

That may be creating an advantage that has very little to do with GPUs.

Every technology company today faces pressure to explain its AI strategy. Whether the company sells networking, security, observability, storage, cloud infrastructure or developer tools, customers want to know how AI fits into the roadmap. Boards want answers. Investors want answers. Customers want answers.

One of the easiest ways to establish credibility is to demonstrate a connection to NVIDIA.

The result is that vendors are effectively opening their customer communities to NVIDIA.

Cisco is hardly alone. Storage vendors discuss how they support NVIDIA-powered AI factories. Security vendors explain how they protect NVIDIA-based environments. Platform vendors showcase their ability to deploy and manage NVIDIA workloads. Infrastructure providers highlight their NVIDIA integrations. Everyone is telling their own story, but NVIDIA’s story often travels alongside it.

From NVIDIA’s perspective, that’s an extraordinarily efficient form of distribution.

Years ago, Intel understood something similar with the Intel Inside campaign. The genius of Intel Inside wasn’t the sticker on the laptop. The genius was convincing hundreds of manufacturers to help market Intel’s brand. Every advertising dollar spent by a PC vendor reinforced Intel’s position. Intel became larger than the individual products it sold because it became part of the broader industry narrative.

NVIDIA appears to be building a 21st century version of that advantage.

The difference is scale.

Intel wanted to be inside every PC. NVIDIA is positioning itself as the foundation underneath nearly every serious AI initiative. Networking vendors want the association. Security vendors want the association. Cloud providers want the association. Enterprise software vendors want the association. The number of technology categories now seeking alignment with NVIDIA is significantly larger than what Intel enjoyed during the PC era.

There are echoes of Cisco in the networking boom and VMware during the virtualization era as well. Both companies reached a point where customers assumed their presence in major projects. If you were building enterprise networks in the late 1990s, Cisco was often the default assumption. During the virtualization boom, VMware occupied a similar position. Competitors existed, but the burden of proof frequently fell on alternatives rather than the market leader.

NVIDIA increasingly appears to occupy that kind of position in AI infrastructure.

What’s remarkable is that the company is strengthening that position while much of the industry is helping it do so.

Every joint announcement extends NVIDIA’s reach.

Every conference appearance extends NVIDIA’s reach.

Every partnership announcement extends NVIDIA’s reach.

Every keynote slide featuring an NVIDIA logo extends NVIDIA’s reach.

Viewed individually, these moments seem insignificant. Viewed collectively, they create something much more powerful: mindshare.

Mindshare is difficult to measure and easy to underestimate. Companies tend to focus on market share because it appears on spreadsheets and earnings reports. Yet many of the most successful technology companies built their dominance long before the numbers reflected it. They first became the default answer in industry conversations.

That’s what I increasingly see happening with NVIDIA.

When practitioners discuss AI infrastructure, NVIDIA often enters the discussion before competitors are even considered. When vendors build AI roadmaps, NVIDIA frequently appears somewhere in the architecture. When conference organizers assemble AI programming, NVIDIA becomes an obvious participant.

None of this guarantees permanent leadership. Technology history is full of dominant companies that eventually lost their position. IBM experienced it. Microsoft experienced it. Intel experienced it. Cisco experienced periods where competitors narrowed what once looked like insurmountable advantages.

The lesson from those examples is not that leadership disappears. The lesson is that leadership must continually be reinforced.

Watching NVIDIA at Cisco Live, I came away thinking the company understands that better than many of its competitors.

The semiconductor industry traditionally obsesses over performance, manufacturing capacity, software ecosystems and technical roadmaps. NVIDIA still enjoys meaningful advantages in all of those areas. What I increasingly believe the company is building, however, is something less tangible and potentially just as valuable.

It is building distribution.

Not distribution through resellers or channel partners in the traditional sense. Distribution of influence. Distribution of attention. Distribution of mindshare across every major technology community that will play a role in the adoption of AI.

Competitors can challenge products. They can challenge pricing. They can challenge performance benchmarks. Challenging a company that has become embedded in the industry’s daily conversation is far more difficult.

That may be the larger lesson from Cisco Live. NVIDIA’s strongest moat may not ultimately be Blackwell, CUDA or whatever comes after them. It may be the fact that so much of the technology industry has concluded that telling its own AI story works better when NVIDIA is part of the narrative.