Wiz stretches cloud security from AI-generated code all the way to internet edge

AI coding tools are making cloud development faster. At the same time, according to Wiz Research, they are making cloud applications measurably less secure. Analysis of real applications built with AI-assisted coding tools found that 20 percent contained significant security issues, including broken access controls and exposed data endpoints. That finding sits at the center of Wiz’s latest cloud security platform expansion, announced alongside Google Cloud Next, which stretches coverage from the moment AI-generated code gets written all the way through to the cloud edge.

To address the coding problem directly, three new functions land inside Wiz Code. First, AI-BOM builds an inventory of the AI frameworks, models, and IDE extensions running across an organization’s cloud coding environment, covering tools like LangChain, Gemini Code Assist, GitHub Copilot, and Cursor. The goal here is reducing shadow AI, where developers reach for coding tools that never went through standard cloud security review. Beyond that, security guardrails scan AI-generated code inside tools like Lovable before it reaches source control, applying organizational cloud security rules at the point of generation rather than catching problems after cloud deployment.

Additionally, pre-built remediation skills for Claude Code and Cursor let AI agents pull validated findings from the Wiz Security Graph and apply fixes directly inside the developer environment.

Beyond code, Wiz extended its cloud security platform into agent studio environments, adding coverage for AWS Agentcore, Gemini Enterprise Agent Platform, Microsoft Azure Copilot Studio, and Salesforce Agentforce. Furthermore, support for Databricks arrives with this update, giving cloud security teams clearer visibility into where sensitive data sits and how identity, infrastructure, and access patterns combine to create exploitable cloud risk.

As part of the same release, Red Agent enters public preview, an AI system that models attacker behavior to identify and validate complex cloud vulnerabilities. It joins Blue Agent, which reached general availability, and Green Agent, currently in public preview, together rounding out the agent-based cloud security tier inside the platform.

Meanwhile, cloud edge infrastructure coverage expands through integrations with Cloudflare, Akamai, Vercel, and Google Cloud Apigee. Those connections pull edge services, DNS configurations, front-end deployments, and API gateways into the Wiz Security Graph alongside cloud workloads, consequently giving security teams a connected view of risk that spans environments most cloud platforms still treat as separate concerns.

For organizations where AI tools now touch every layer of cloud development and deployment, that connected view is increasingly becoming a basic operational requirement rather than an optional upgrade.