Microsoft launches preview of Security Dashboard for AI
Microsoft has released the preview of the Security Dashboard for AI. It provides an overview of AI risks in the company.
(Image: Microsoft)
Microsoft has released the public preview of the "Security Dashboard for AI". It is intended to provide IT managers with a simple overview of the risks posed by AI assistants, apps, and platforms used.
In a blog post, Microsoft introduces the Security Dashboard for AI. It was already announced at the Ignite trade fair 2025, where Microsoft had set an AI focus. With the Security Dashboard, the company wants to provide IT security personnel with a clear overview of AI risks such as data leaks, vulnerabilities in models, misconfigurations, and "unethical agent actions" that encompass the entire AI environment, AI platforms, apps, and agents.
"The Security Dashboard for AI provides executives with a unified view of AI risks by consolidating security, identity, and data risks from Defender, Entra, and Purview into a single interactive dashboard," explains Microsoft. The dashboard's overview tab provides a scoring overview with cards that immediately show where risks may lie that IT security teams need to address. It also captures the implementation of Microsoft's Security-for-AI capabilities within the organization and provides recommendations for improving AI security.
AI Inventory
The dashboard also includes an AI inventory with comprehensive views for finding AI usage, risk assessments, and countermeasures with broad coverage of AI agents, models, MCP servers, and apps. The Security Dashboard for AI can handle all Microsoft AI solutions supported by Microsoft's Entra, Defender, and Purview, including Microsoft 365 Copilot, Copilot Studio agents, and Foundry applications and agents. It also recognizes third-party AI models, applications, and agents such as Google Gemini, OpenAI ChatGPT, and MCP servers.
Videos by heise
Microsoft promises that the dashboard provides a summary overview of AI-related security risks and uses AI-powered insights from Security Copilot to identify the most critical risks within an environment. This allows IT security teams to quickly identify and address the most pressing risks. And, of course, IT security officers can also use AI prompts to go through the discovered risks and analyze them more precisely.
Microsoft Security customers can start using the preview version of the Security Dashboard for AI directly. It is included in the license for Microsoft Defender, Entra, and Purview.
(dmk)