Cisco Launches AI Defense Hybrid Model Across 3 Major Clouds and On-Premises
Updated
Updated · Cisco Blogs · Jun 2
Cisco Launches AI Defense Hybrid Model Across 3 Major Clouds and On-Premises
3 articles · Updated · Cisco Blogs · Jun 2
Cisco said AI Defense now runs as a pure software layer on AWS, Microsoft Azure, Google Cloud Platform and on-premises through Cisco Secure AI Factory with NVIDIA.
The hybrid model is meant to remove infrastructure lock-in as enterprises spread AI workloads across multiple clouds, GPU types and deployment targets while facing a wider attack surface from agentic systems.
Three security functions travel with the application: supply-chain scans for models, datasets and MCP servers; runtime guardrails on requests and responses; and agent-to-agent protection using NVIDIA OpenShell.
Deployment is Kubernetes-native with validated support for EKS, AKS, GKE and Red Hat OpenShift, letting customers extend from a 4-node AWS cluster to Azure regions and on-premises without re-architecting policies.
Does a universal security approach sacrifice performance for compatibility on specialized AI cloud platforms?
Is Cisco's platform a silver bullet for EU AI Act compliance, or just another complex security layer?
Can security guardrails protect an autonomous AI workforce without crippling its problem-solving capabilities?
2026 AI Security Report: Cisco’s Unified Defense Tackles 69% Spike in AI Vulnerabilities
Overview
The rapid evolution and widespread adoption of artificial intelligence have created unprecedented security challenges, leading to an urgent need for robust defense mechanisms. As of June 2026, AI security faces an alarming rise in vulnerabilities and exploits, with vulnerability exploitation becoming the main cause of cyberattacks in 2025. This trend is intensifying, as research projects a dramatic increase in AI-related vulnerabilities for 2026. The shrinking window between vulnerability discovery and exploitation highlights the need for immediate, proactive security responses. These developments underscore the critical importance of comprehensive solutions to protect AI applications and infrastructure in an increasingly complex threat landscape.