Microsoft Enhances Zero Trust Framework for AI Security

• •Microsoft has introduced a 91-task DevSecOps pillar to its Zero Trust for AI framework.
• •The Zero Trust Assessment tool now includes an AI-specific evaluation component.
• •Organizations can use these updates to improve their security posture regarding AI technologies.

Microsoft has expanded its Zero Trust for AI strategy, introducing updates to the Zero Trust Assessment tool and a new 91-task DevSecOps pillar. These enhancements are designed to help organizations evaluate their security posture, prioritize remediation efforts, and apply zero trust principles to AI agents and AI-assisted software development. The Zero Trust Assessment tool evaluates Microsoft security configurations against best practices and identifies weaknesses. The updates aim to provide security teams with a structured approach to baseline AI-agent risks and develop remediation roadmaps. This initiative is part of a broader effort to secure AI technologies within enterprise environments. The updates were announced on August 6, 2026, and are available to organizations using Microsoft services.