A security engineer is configuring Microsoft Purview to automatically apply sensitivity labels to Azure OpenAI-generated reports stored in SharePoint Online. The reports contain financial forecasts. The engineer creates an auto-labeling policy with a trainable classifier for financial reports. The policy has been active for 14 days but no labels have been applied. What is the most likely reason?

The correct reason is that the auto-labeling policy is in simulation mode and must be turned on after reviewing simulation results. Microsoft Purview auto-labeling policies for SharePoint and OneDrive start in simulation mode by default. In simulation mode, the policy identifies what would be labeled without actually applying labels. After reviewing the simulation results, administrators must turn the policy on, or active mode, to begin applying labels. If the engineer did not activate the policy after simulation, labels will not be applied even after 14 days. The claim that trainable classifiers require a minimum of 50 positive samples to activate in auto-labeling policies is incorrect because trainable classifiers have training data requirements during the creation and training phase, not a minimum sample requirement before auto-labeling policies can use them. The claim that auto-labeling policies for SharePoint require the Microsoft 365 E5 Compliance license, which may not be assigned, is incorrect because while sensitivity label auto-labeling does require appropriate licensing, the most common cause of no labels being applied is simulation mode still being active. The claim that the SharePoint site must be added to the Purview Data Map before auto-labeling can apply labels is incorrect because SharePoint sites do not need to be registered in Purview Data Map for auto-labeling policies to apply sensitivity labels to SharePoint content.