Enterprise AI adoption isn’t slowing down — and neither are the dangers that include it. Based on the 2025 Cisco Cybersecurity Readiness Index, 86% of organizations skilled an AI-related safety incident prior to now 12 months, but fewer than half have the assets to conduct complete AI safety assessments.
Cisco AI Protection was constructed to assist shut that hole. At the moment, we’re extending that safety to Google Cloud — becoming a member of AWS and Microsoft Azure because the third main cloud platform that Cisco AI Protection helps.
What This Means for Google Cloud Clients
Whether or not you’re operating fashions on Gemini Enterprise Agent Platform, constructing agentic workflows, or managing delicate information as a part of a RAG pipeline, AI Protection now delivers enterprise-grade safety inside your Google Cloud surroundings.
Key capabilities embody:
AI Mannequin & Software Validation — Automated red-teaming throughout 200+ safety and security subcategories mapped to the Cisco AI Safety and Security Framework together with OWASP, NIST, and MITRE requirements
AI Runtime Safety — Bi-directional guardrails enforced inline throughout agentic workloads, overlaying threats like immediate injection, software misuse, and information exfiltration — with no code modifications required
AI Cloud Visibility — Automated discovery of AI belongings together with fashions, brokers, information sources, and tooling, offering centralized governance
AI Provide Chain Danger Administration — Steady scanning of mannequin information, repositories, and MCP servers to determine vulnerabilities and forestall the usage of insecure elements
For organizations with strict information necessities, a VPC deployment choice retains all information inside your Google Cloud surroundings — no exterior routing of prompts, responses, or mannequin interactions.
Constructed for Agentic AI
As enterprises transfer from AI experimentation to production-scale agent deployments, runtime safety turns into essential. AI Protection integrates instantly with Google Cloud’s Agent Gateway through Google Kubernetes Engine (GKE) Service Extensions, working as an inline coverage enforcement engine on agent requests and responses — with out touching agent code or mannequin configuration.
For builders, getting began is simple. AI Protection is accessible as an ADK integration, permitting groups to allow runtime safety with only a few strains of code when constructing on Gemini Enterprise Agent Platform Runtime .
Constant Safety Throughout Your Multi-Cloud AI Footprint
For enterprises constructing AI throughout AWS, Azure, and Google Cloud, AI Protection gives a single, constant safety framework — the identical visibility, the identical controls, the identical governance — no matter the place your AI runs. AI Protection maps to OWASP Prime 10 for LLMs, MITRE ATLAS, and NIST Adversarial ML Taxonomy, supporting compliance readiness together with the EU AI Act.
Visiting Google Cloud Subsequent 2026 in Las Vegas? Cease by Cisco Sales space #4813 to see AI Protection in motion. Or contact your Cisco account consultant to study extra.
