Outlook Microsoft 365 Encoding Issue (February 2026): Why It Suddenly Broke and How to Fix It

On February, 2026, organizations running Microsoft 365 Apps for Enterprise began experiencing a sudden encoding issue in Outlook:

• Special characters (such as typographic quotes « ») were replaced with ?
• HTML emails were corrupted
• The issue did not affect Office 2016
• Everything was working fine just two days earlier

The affected build:

Version 2602 (Build 16.0.19725.20126) 64-bit

This article explains what changed, why it happened, and how to permanently fix it.

Fix: Create a Group Policy to Enforce UTF-8

If you need to resolve the issue quickly in a production environment, apply the following Group Policy configuration.

This forces Outlook to use UTF-8 for outgoing HTML messages and immediately stops character corruption.

Open Group Policy Management

1. Open Group Policy Management Console (GPMC).
2. Create a new GPO (for example: Outlook – Force UTF-8 Encoding).
3. Link it to the OU containing affected users.

Important: This is a User Configuration policy.

Create the Registry Item

Navigate to:

User Configuration → Preferences → Windows Settings → Registry

Create a new Registry Item with the following settings:

Action: Update
Hive: HKEY_CURRENT_USER
Key Path:

Software\Microsoft\Office\16.0\Outlook\Options\MSHTML\International

Value name:

Default_CodePageOut

Value data (Hex):

0000FDE9

0000FDE9 (hex) = 65001 (UTF-8)

Apply the Policy

• Run: gpupdate /force
• Completely close Outlook (including background processes).
• Reopen Outlook.

Verify the Fix

Send a test email and check the headers:

File → Properties → Internet Headers

You should now see:

Content-Type: text/html; charset=utf-8

If UTF-8 is present, the issue is resolved.

What Actually Happened

The issue was caused by a change in how Outlook handles HTML encoding in newer Microsoft 365 builds.

Starting with build 16.0.197xx, Microsoft modified:

• The MIME header generation logic
• Charset selection behavior for HTML messages
• MSHTML handling of legacy code pages

If Outlook was configured to use a legacy encoding such as KOI8-R, the new engine began strictly enforcing that encoding for outgoing HTML messages.

Previously, Outlook would often fall back to UTF-8 internally, even if another code page was configured. After the update, that fallback behavior appears to have been removed or tightened.

As a result:

• Characters not supported by KOI8-R were replaced with ?
• HTML messages declared charset=koi8-r
• Non-ASCII symbols became corrupted