Wiz Launches AI-APP: Google Cloud's First Post-Acquisition Platform for Securing AI Applications End-to-End
At RSAC 2026 on March 23, Wiz — now part of Google Cloud following the $32 billion acquisition — launched the AI Application Protection Platform (AI-APP), a new security category designed to protect every layer of AI applications from code to runtime.
AI-APP treats AI as a dynamic, interconnected system rather than a collection of isolated components. The platform provides full visibility into models, agents, and data flows across AWS Bedrock, Azure AI, and Vertex AI, mapping complex attack paths and correlating cloud-native risks in a single graph-powered platform. It protects against prompt injection, shadow AI, and the unique risks that emerge when agents autonomously access data and tools.
The platform introduces a three-agent architecture: Green Agents that continuously assess security posture, Blue Agents that respond to incidents, and the new Red Agent that simulates real attacker behavior to find vulnerabilities before they're exploited. Together they form an agentic security workflow where AI defends AI.
Google Cloud simultaneously announced its broader agentic AI defense strategy with Wiz integration, combining Mandiant threat intelligence with Wiz's cloud security graph for unified AI application protection.
For the agentic ecosystem, AI-APP represents a shift from protecting individual AI models to securing the entire AI application stack — infrastructure, data, access, models, agents, and applications — as a single interconnected surface. This is the security model that agent-dense enterprises will need.
https://www.wiz.io/blog/introducing-wiz-ai-app
← Back to all articles
AI-APP treats AI as a dynamic, interconnected system rather than a collection of isolated components. The platform provides full visibility into models, agents, and data flows across AWS Bedrock, Azure AI, and Vertex AI, mapping complex attack paths and correlating cloud-native risks in a single graph-powered platform. It protects against prompt injection, shadow AI, and the unique risks that emerge when agents autonomously access data and tools.
The platform introduces a three-agent architecture: Green Agents that continuously assess security posture, Blue Agents that respond to incidents, and the new Red Agent that simulates real attacker behavior to find vulnerabilities before they're exploited. Together they form an agentic security workflow where AI defends AI.
Google Cloud simultaneously announced its broader agentic AI defense strategy with Wiz integration, combining Mandiant threat intelligence with Wiz's cloud security graph for unified AI application protection.
For the agentic ecosystem, AI-APP represents a shift from protecting individual AI models to securing the entire AI application stack — infrastructure, data, access, models, agents, and applications — as a single interconnected surface. This is the security model that agent-dense enterprises will need.
https://www.wiz.io/blog/introducing-wiz-ai-app
Comments