Is Cisco's Cybersecurity Architecture Ready for Agentic AI?
by James Darley, Technology Magazine
Cisco has announced a comprehensive overhaul of its security infrastructure designed to address the emerging challenges posed by autonomous AI agents and AI-powered cyber threats.
At Cisco Live in San Diego, executives from the networking giant outlined how enterprises are struggling to balance the rapid adoption of AI with the increased security concerns that it brings in tow.
While AI has the potential to help improve cybersecurity systems by automating threat detection and triggering faster responses, the technology is equally as useful for threat actors, many of whom are leveraging it to launch more sophisticated attacks.
"Safety and security are the defining challenges of the AI era — and agentic AI multiplies the risk, as every new agent is both a force multiplier and a fresh attack surface," says Jeetu Patel, President and Chief Product Officer at Cisco.
The company's response centres on deeper integration of security capabilities into networking infrastructure, moving away from traditional perimeter-based approaches towards what it calls "zero trust architectures" that continuously verify all network participants.
A new era of hardware
Central to Cisco's strategy is an expansion to its Hybrid Mesh Firewall portfolio, which will feature two new hardware series designed for different scenarios.
The Secure Firewall 6100 Series targets AI-ready data centres with performance density of 200 Gbps per rack unit, addressing what Cisco identifies as complexity and scalability challenges in high-performance computing environments.
For distributed branch offices, the Secure Firewall 200 Series integrates threat inspection with software-defined wide area networking, claiming up to three times better price-performance compared to competitors.
This hardware refresh accompanies expanded policy management capabilities through Cisco's Security Cloud Control platform, which will extend unified management to next-generation firewalls across the company's SD-WAN, switching and data centre fabric products.
The unique challenge of agentic AI
The emergence of agentic AI is presenting a particular challenge for security teams across the global economy.
The autonomous nature of these systems means that they can access resources, make decisions and act independently on behalf of teams — a good thing for efficiency, but a potential vulnerability for cybersecurity.
Cisco has a new offering known as 'Universal Zero Trust Network Access' which aims to address this vulnerability. It provides identity-driven access controls that extend to both human users and AI agents across hybrid environments.
The approach integrates automated agent discovery, delegated authorisation and comprehensive tracking of agent actions through what Cisco describes as native support for the Model Context Protocol.
Splunk's role in Cisco's cybersecurity push
Since Cisco acquired data analytics platform Splunk in early 2024, it has been able to refine and improve its security systems like never before, enabling what the company describes as "advanced detections" and helping security teams extract greater value from their existing investments.
Additional SOAR integrations now include Cisco Secure Firewall-specific actions for automated threat containment, including host isolation and connection blocking capabilities.
"As AI continues to evolve at an unprecedented pace and new cybersecurity challenges emerge, it's even more important to fuse security into the very fabric of the network," explains Chris Konrad, Vice President of Global Cyber at World Wide Technology.