Microsoft Security Copilot is a new GPT-4-based AI cybersecurity assistant

Microsoft announced a new product for cybersecurity professionals – Microsoft Security Copilot. It is an AI assistant that uses OpenAI GPT-4 generative artificial intelligence and Microsoft’s proprietary security model to help professionals detect and stop cyberattacks.

Security Copilot allows security professionals to ask questions in natural language and get answers based on the 65 trillion daily signals that Microsoft collects from a variety of sources. Security Copilot also integrates with other Microsoft security tools such as Sentinel, Defender, and Intune to provide recommendations for mitigating threats and improving security.

Security Copilot does not replace security professionals, but complements their work. It also includes a section for collaborating and sharing information between colleagues. It is noted that Security Copilot does not use user data to train other AI models.

One of the most interesting features of Security Copilot is the ability to create a hint directory. It’s basically a set of steps or automatic actions that people can combine into one easy-to-use button or tooltip. For example, you can create a generic script reengineering hint so that security researchers don’t have to wait for someone on the team to do this analysis. You can even use Security Copilot to create a PowerPoint slide that describes incidents and attack vectors.

Security Copilot is in preview and is only available to select Microsoft enterprise customers.