MC1301831 - Microsoft Purview: Data Security Investigations – Introducing optical character recognition (OCR) support

Introduction

Microsoft Purview Data Security Investigations (DSI) is expanding its AI-powered investigation capabilities by adding optical character recognition (OCR). This enhancement enables DSI to extract and analyze text from images, helping organizations identify sensitive information embedded in visual content. This improves the accuracy and depth of data security investigations.

This message is associated with Microsoft 365 Roadmap ID 561489.

When this will happen:

Public Preview (Worldwide): We will begin rolling out in late May 2026 and expect to complete by early June 2026.
General Availability (Worldwide): We will begin rolling out in mid-July 2026 and expect to complete by late July 2026.

How this affects your organization:

Who is affected:

Admins and analysts using Microsoft Purview Data Security Investigations (DSI)
Organizations investigating data security risks using Purview

What will happen:

OCR will be enabled by default in Data Security Investigations.
DSI will automatically extract text from image-based content (for example: images, screenshots, embedded visuals in files).
Extracted text will be incorporated into investigation datasets to improve search, analysis, and risk detection.
Existing investigation workflows will require no changes.
This can help improve detection of sensitive information that may be embedded in visual content.
Existing Purview policies and controls (such as sensitivity labels and DLP) continue to be respected.

What you can do to prepare:

No action is required prior to rollout.

You may consider the following:

Inform your security and compliance teams about improved detection capabilities involving image-based content.
Review internal investigation procedures to account for insights derived from OCR.
Update any internal documentation or training materials that reference Data Security Investigations capabilities.

Learn more:

Compliance considerations:

Consideration	Explanation
Alters how existing customer data is processed	OCR introduces additional processing of image-based content in Data Security Investigations by extracting text for analysis.
Introduces or modifies AI/ML capabilities	AI-powered OCR is added to analyze visual content and enhance investigation insights.
Alters admin monitoring, reporting, or compliance visibility	OCR-enriched data improves investigation depth, which may impact reporting and how compliance activities are reviewed.