Cisco Expands Security Portfolio to Protect Agentic AI in Enterprises
Cisco has announced a significant evolution of its security portfolio aimed at assisting enterprises in adopting agentic AI with confidence. This expansion integrates agent protection, interaction governance, and resilient connectivity to support AI-driven workflows.
The Need for Enhanced Security
As organizations transition from simple AI assistants to more autonomous agents that utilize tools and data across hybrid environments, security teams are compelled to bolster agentic defenses. Effective governance over agent interactions with enterprise systems and external services is crucial, alongside maintaining reliable, cryptographically protected connectivity at scale.
Jeetu Patel, President and Chief Product Officer at Cisco, emphasized, “In the age of AI, safety and security are prerequisites for adoption, and AI agents bring a whole new set of challenges. As agents take on critical enterprise roles, we’re developing protections that work both ways: preventing agents from being compromised and controlling what they can access and do on our behalf.”
Protecting Agents from Compromise
The rise of agentic AI has widened the attack surface across AI supply chains and the tool ecosystem. Therefore, enterprises require robust protections to mitigate the risks of agents being manipulated or hijacked during tool interactions.
Cisco’s latest expansion, the AI Defense, introduces new features designed to enhance the security of agents and the AI supply chain:
- AI BOM (Bill of Materials): Offers centralized visibility and governance for AI software assets, ensuring the security of the AI supply chain.
- MCP Catalog: Discovers and manages risks across model context protocol (MCP) servers and registries, reinforcing AI governance.
- Advanced algorithmic red teaming: Extends the scope of AI security assessments through adaptive testing for models and agents in multiple languages.
- Real-time agentic guardrails: Continuously monitors agent interactions to detect manipulation or unsafe behavior.
Governing Agent Interactions
AI agents depend on ongoing interactions with large language models (LLMs), SaaS applications, data stores, and tool endpoints, often located remotely. To address these needs, Cisco’s SASE is launching new capabilities designed to govern agent interactions and ensure reliability in AI traffic:
- AI traffic optimization: Employs techniques such as packet duplication to maintain low-latency AI interactions during traffic surges.
- MCP visibility and control: Utilizes in-path controls to manage agent-to-tool connectivity effectively.
- Intent-aware inspection: Combines rapid detection techniques with cloud-based analysis to assess the intent behind agentic messages and actions.
- Unified policy enforcement: Coordinates controls within a single framework to streamline governance as agent adoption accelerates.
Delivering Reliable Connectivity
To tackle the challenge of reliable connectivity, Cisco has introduced IOS XE 26, the latest version of the operating system that powers millions of networks worldwide. This update supports the recently announced Cisco 8000 Series Secure Routers and Cisco C9000 Series Smart Switches.
IOS XE 26 features industry-first full-stack post-quantum cryptography (PQC) protections tailored for enterprises. This enhancement aims to defend organizations against device tampering and data compromise, aligning with evolving European and global regulatory standards.
In summary, Cisco’s expansion of its security portfolio signifies a proactive approach to securing agentic AI, ensuring that enterprises can adopt these technologies confidently while mitigating the associated risks.
