At the Google Cloud Security Summit, Google shared updates on helping organizations build securely, secure systems, and defend at machine speed in the agentic era, stating that passive defense is no longer viable and that ungoverned adoption of autonomous systems into traditional workflows can lead to critical security incidents. Google said agentic coding tools have accelerated software delivery while human review can’t keep up with machine-speed commits and unvetted dependencies, and that because autonomous agents are now part of the supply chain, blind trust can turn them into direct backdoors for attackers. Its approach builds native, real-time guardrails in the integrated development environment (IDE), unifies asset graphs across code and identity and access management (IAM), and verifies models to weed out logic flaws before code merges. Google described CodeMender, an autonomous code-security agent that helps find and fix code vulnerabilities at machine speed, eliminates manual triage bottlenecks, and works with multiple models including Gemini 4 Argon. Google said Wiz used Gemini 4 Argon along with other models to identify and help remediate core PHP vulnerabilities that affected 70% of websites. CodeMender, in preview, adds secure autonomous remediation using an operating system-level sandbox or isolated virtual machine with human-in-the-loop checks for local write and command execution, and flexible scanning modes for a specific folder, entire repository, or only modified files since the last scan. Since the CodeMender preview became available in July, hundreds of customers have used it