Microsoft Targets Rogue AI Agents With Dynamic Permissions, Risk Scoring

Microsoft is introducing controls to manage AI agents, treating them as insiders within companies. These controls will grant access to AI agents on a per-task basis and will tighten access as the agent's risk level increases. The goal is to mitigate the potential risks associated with rogue AI agents. Microsoft's approach involves dynamic permissions and risk scoring to ensure AI agents' access is appropriately restricted. This method allows for more fine-grained control over AI agents' activities. Companies can use this approach to better manage the security risks posed by AI agents.