Over the past year, security leaders across various disciplines have been rapidly adopting AI. However, this has raised concerns about whether traditional security models can keep pace. Microsoft is addressing these issues by introducing a new approach called Zero Trust for AI (ZT4AI). This extends proven Zero Trust principles to all stages of the AI lifecycle, from data ingestion and model training through deployment and agent behavior.
The updated Microsoft Zero Trust Workshop now includes an AI pillar, alongside enhanced Data and Networking pillars. Additionally, there is a new Zero Trust reference architecture for AI and practical patterns for securing AI at scale. These tools aim to provide a clear path from understanding what needs to be done to taking action.