Check Point Software Technologies Ltd. announced it is collaborating with Microsoft to deliver enterprise-grade AI security for Microsoft Copilot Studio. The collaboration enables enterprises to safely build and deploy generative-AI agents with continuous protection, compliance, and governance integrated directly into their development workflows. The integration with Copilot Studio brings together Check Point's AI Guardrails, Data Loss Prevention (DLP), and Threat Prevention technologies, extending its end-to-end AI security stack to safeguard Copilot Studio during agent runtime.
The result is continuous protection for every AI agent, ensuring safe and compliant innovation. As enterprises rapidly adopt AI agents to drive productivity, new risks emerge, from prompt injection and data leakage, to model misuse and compliance drift. These agents connect to sensitive data and third-party tools, expanding the attack surface beyond traditional controls.
By using Check Point's runtime security and governance capabilities to extend Copilot Studio's protections, organizations gain full visibility and control to innovate confidently and securely. Key capabilities include: runtime AI Guardrails - Continuous runtime protection for every agent built with Copilot Studio, preventing prompt injection, data leakage, and model misuse; Data Loss and Threat Prevention - Integrated DLP and Threat Prevention engines that safeguard sensitive data across every tool call and workflow inside Copilot Studio; Enterprise-Grade Scale and Precision - A unified security bundle designed for large-scale deployments, delivering consistent protection and low latency without impacting performance; Seamless Protection for Productivity - Allows organizations to fully use the power of Copilot Studio while maintaining runtime visibility, compliance, and prevention-first protection. This collaboration reinforces Check Point's leadership in securing the AI-powered enterprise and marks another milestone in its mission to protect the full AI lifecycle - from model development to runtime execution, and from organizational applications to employee usage across the workspace.
