CodeMender Adds Finer Scan Controls for Security Teams
The AI vulnerability tool remains in preview, so broad availability and pricing have not been announced.
Updated on Nov. 5, 2026 in Cybersecurity
Live Poll
Do you trust AI tools to find and fix security flaws in software you use?
Google Cloud added new controls to CodeMender, its AI tool for finding and fixing software vulnerabilities, and extended security and governance controls across environments where AI agents run. CodeMender remains in preview, so these capabilities are not yet broadly available.
Why it matters
The additions push AI-driven vulnerability discovery toward enterprise use by letting teams control where and how scans run. Extending governance to autonomous agents reflects how cloud security is being rebuilt around software that acts on its own.
A Thales evaluation put the potential reduction in vulnerability discovery time at 95%, but the source does not state the baseline or test conditions behind that figure. Treat it as an early vendor-cited estimate rather than an established benchmark.
The players
Google Cloud
Google's cloud division, whose stack spans AI services, identity management, and network security controls for enterprise customers.
Thales
A global security and data-protection company whose evaluation cited the 95% potential reduction in vulnerability discovery time.
The details
CodeMender lets businesses isolate attack simulations and narrow code scans to specific targets. It can verify a vulnerability inside an operating system-level sandbox or an isolated virtual machine, a contained environment that keeps tests from touching production systems. Businesses can scan a full code repository, a single folder, or only files changed since the last scan. Human checks remain in place for local writes and command execution, so a reviewer must approve actions that modify files or run commands. Separately, Auth Manager centralizes credential management in a Google-managed vault governed by identity and access management (IAM), the system that controls who can access which cloud resources.
Timeline
Google Cloud's CodeMender controls and related security updates were reported on November 5, 2026.
The Tech Race
The CodeMender update places Google Cloud alongside a broader push by cloud providers to tie AI agents to the same identity and access controls used for human users. Extending IAM-governed credential management to autonomous agents marks a shift from securing people's accounts toward securing software that operates without a person at each step.
Security teams can choose whether to scan a full repository, one folder, or only recently changed files, and can isolate tests from live systems. CodeMender remains in preview, and the source gives no general availability date or pricing, so timing for broad access is unannounced.
The takeaway
Watch whether CodeMender moves from preview to general availability and whether the 95% discovery-time figure is reproduced in independent testing. Both would show whether AI-driven vulnerability discovery is ready for production use.
Further reading
For more on how AI tools are reshaping threat detection and software defense, explore our Cybersecurity coverage.
Source note: This article includes information reported by Help Net Security.
Live Poll
Do you trust AI tools to find and fix security flaws in software you use?
