Wiz Introduces AI Security Agent for Assessing AI Applications Across Code and Cloud

Summary
Wiz says its AI Security Agent, now in Public Preview, inventories AI apps across code, PaaS and self-hosted environments, analyzes agent capabilities, and links application risks to cloud exposure, permissions and data in the Wiz Security Graph.
Key points
- Wiz announced the AI Security Agent in Public Preview, extending its cloud security platform to analyze AI application logic, prompts and configurations.
- The product is designed to inventory custom AI applications, models, prompts, MCP servers and tool definitions across code repositories, supported PaaS platforms and self-hosted environments.
- Wiz says the agent compares intended purpose with actual capabilities and permissions, highlighting issues such as excess access or unintended external exposure.
- The product correlates application findings with cloud context—including internet exposure, identities, permissions and sensitive data—to map potential attack paths.
- Wiz illustrates the approach with an example involving an unvalidated endpoint, exposed credentials and sensitive data; the article presents this as a product scenario, not a reported incident.
- Wiz says it has detected up to five times as many AI agents as traditional discovery engines across customer environments; this is a vendor-reported claim.
Article Details
- Event Type
- Wiz announced the Wiz AI Security Agent, now available in Public Preview.
- Impact
- The agent analyzes AI applications across self-hosted environments, supported PaaS platforms, and code repositories. It combines application-level findings with cloud context to identify agents, assess their purpose and capabilities, surface AI-native risks, and map potential attack paths involving exposure, permissions, and sensitive data. Wiz says it detected up to five times as many AI agents as traditional discovery engines across customer environments.
Vendors
Products
Wiz AI Security AgentTake, for example, a hypothetical BigQuery Agent. Reviewing its code, the Wiz AI Security Agent identifies tools with read and write access to multiple BigQuery databases, along with an arbitrary code executionWiz Security Graphwith how it’s configured at the cloud infrastructure layer. By feeding these insights into the Wiz Security Graph, the AI Security Agent evaluates the vulnerability against broader cloud context, checking for
