Microsoft Treats AI Agents as Insider Threats in New Security Framework

Microsoft urges organizations to secure AI agents as insider threats, using dynamic permissions, continuous risk monitoring and automated containment, Security Affairs reports. The company published detailed guidance explaining how organizations should secure agents as their access and autonomy grow. SiliconANGLE’s version of the same announcement was dropped as a duplicate.