MC1217615 – New Content Moderation Settings for AI Prompts in Microsoft Copilot Studio and Power Platform

Product:

Copilot, Power Platform

Platform:

Online, World tenant

Status:

Change type:

Feature update

Links:

Details:

We are announcing the ability to configure content moderation levels for AI prompts in Microsoft Copilot Studio and Power Platform. This feature setting will reach general availability on January 31, 2026.

Change Category:
XXXXXXX ... free basic plan only

Scope:
XXXXXXX ... free basic plan only

Release Phase:

Created:
2026-01-14

updated:
2026-01-14

Task Type

XXXXXXX ... free basic plan only

Docu to Check

XXXXXXX ... free basic plan only

MS How does it affect me

XXXXXXX ... free basic plan only

MS Preperations

XXXXXXX ... free basic plan only

MS Urgency

XXXXXXX ... free basic plan only

MS workload name

XXXXXXX ... free basic plan only

summary for non-techies**

XXXXXXX ... free basic plan only

Direct effects for Operations**

Increased Risk of Harmful Content
Lowering content moderation levels may lead to the generation of harmful or inappropriate content, negatively affecting user trust and safety.
   - roles: Content Moderator, End User
   - references: https://aka.ms/13595/knowledge-copilot-studio

Reduced Response Quality
Higher moderation settings may limit the variety and creativity of AI responses, potentially frustrating users who seek diverse outputs.
   - roles: Content Creator, End User
   - references: https://aka.ms/13595/knowledge-copilot-studio

Configutation Options**

XXXXXXX ... paid membership only

explanation for non-techies**

Microsoft is introducing a new feature that allows users to adjust content moderation levels for AI prompts in Microsoft Copilot Studio and Power Platform. Think of this feature like a filter you might use on your email inbox. Just as you can set your email filter to catch more or fewer spam messages, you can now set how strictly the AI checks the content of its responses.

When you set the moderation level higher, it's like having a very cautious spam filter. It reduces the risk of getting inappropriate or harmful content, but it might also mean you get fewer messages or responses because some safe ones might be caught in the filter. On the other hand, lowering the moderation level is like loosening the spam filter. You might get more responses, but there's a higher chance some of them might not be suitable.

This feature will be generally available on January 31, 2026, and can be adjusted through the settings panel in the prompt builder. It's designed to give users more control over the kind of content they receive from AI prompts, ensuring it aligns with their needs and preferences. No immediate action is required, but if you're interested in exploring this feature further, you can find more information through the provided link.