Traffic patterns are shifting toward massive uploads, autonomous systems are emerging, and infrastructure is being rethought from the core to the edge. What used to take months to deploy now happens in minutes, forcing service providers to rethink their role in a rapidly evolving ecosystem.
At the same time, security and observability are being redefined in real time. From AI-ready firewalls to data pipelines feeding platforms like Splunk, the ability to detect, respond, and even shut down threats instantly is becoming essential. Yet even as automation accelerates, human expertise remains critical to decision-making. The result is a new balance between speed, intelligence, and control - one that will define how AI scales across industries.
Samir Parikh:
AI is fundamentally changing the traffic patterns in the network. It's no longer about downloading cat videos. The networks have to adapt for that new traffic pattern.
Matt Price:
We're seeing a shift to much, much more upload traffic, especially as service providers have to start dealing with physical and agentic AI. So I think service providers have an enormous role to play now in AI.
Jessica Oppenheimer:
The screens that you see are real firewall data coming from the latest firewall, the 6160, which is built for AI-ready data centers. So how to protect those data centers from attacks. All that data is going into Splunk as a platform.
Greg Owens:
What Cisco's been doing for the last year and a half is fundamentally changing that infrastructure for AI traffic versus traditional data traffic.
Matt Price:
Operating at the speed of AI is really, really challenging. So with our mobility services platform, we're providing a programmable API-first network to allow them to compress things that may have taken months in the past down to minutes today.
Greg Owens:
Moving stuff to the edge is about making sure that the content that's been built is as close to the user as possible, and so that when that traffic is flowing, it has less distance to go.
Samir Parikh:
We're going to see this huge movement around autonomous networking driven by AI because it's critical for the networks to be run by AI to be able to support AI as an application end to end.
Jessica Oppenheimer:
Luckily, we have Splunk as the platform. It's designed for that big data. Up there, we see the different attackers, where they're coming in on the firewall. If there was an AI model that was causing problems, you can shut it down. And if they need to, it can escalate it up to enterprise security with just a click.
Greg Owens:
And we're not at the point where we're doing complete automation where the robots are doing everything. You still rely on the expertise of the human to make the final decision. It just increases efficiency very dramatically, and that's the fundamental change that we see.
Jessica Oppenheimer:
Coolest thing is the new 6160. It's the first time that has been installed in a customer environment. And so here we are at the biggest conference in the world, and it's here protecting the network, and all that data is going into Splunk.
Greg Owens:
But we also have our software side and you can start having inferences or you can have holograms, you can have touchscreens. When people can ask questions, see that really happening in the world of AI with edge. If that equipment's actually in the store, you're going to have a really amazing experience that's going to be customized for that region.
Matt Price:
We're seeing the pace in AI growing exponentially, and we're seeing the opportunities for the networks to leverage that and to provide better, more secure, more reliable services just grow and grow and grow.