web
You’re offline. This is a read only version of the page.
close
Skip to main content

Announcements

News and Announcements icon
Community site session details

Community site session details

Session Id :
Power Platform Community / Forums / Copilot Studio / Chat agent citations/r...
Copilot Studio
Suggested Answer

Chat agent citations/references

(0) ShareShare
ReportReport
Posted on by
Hello,
 
I've created a agent to use for a chatbot on our websites. I can get many things functioning properly but I'm stuck on citations and references showing up. When I leave generative AI orchestration turned on it continues to show citations and references. I can shut this off but then would have to determine all action items based on topics. Does any one have an solutions they can recommend to get these turned off? Thanks you in advance
Categories:
I have the same question (0)
  • Suggested answer
    Sayali-MSFT Profile Picture
    Microsoft Employee on at
    Hello,

    When generative orchestration is enabled, citations are a platform-level behavior and cannot be truly disabled today. This is because the system uses a RAG pipeline (retrieve → ground → generate), and any grounded response automatically emits citations after generation, regardless of prompt instructions or formatting overrides. Citations are treated as a trust and safety feature, not a UI preference.

    Attempts that do not work include prompt engineering, custom system messages, response formatting rules, post-generation trimming inside Copilot Studio, or UI toggles—if orchestration and grounding are on, citations will appear.
    What does work are three practical options:

     

    1. Post-process responses (most common and recommended for website bots): Keep orchestration and grounding on, but strip or hide citations at render time (e.g., ignore citation metadata or hide reference blocks in the UI). This preserves grounded answers and avoids rebuilding orchestration.

    2. Disable grounding only: With no sources, there are no citations—but answers rely solely on the model’s internal knowledge, reducing accuracy and freshness.

    3. Turn off generative orchestration: Removes citations entirely but forces topic-based logic, increases maintenance, and scales poorly for open-ended questions.

    4. Advanced teams often use a hybrid approach: generative orchestration for knowledge Q&A, classic topics for transactional flows, and frontend suppression of citations to maintain a clean UX.
    There’s no “Disable citations” toggle yet because citations are tied to Responsible AI commitments, especially for enterprise and regulated data. A toggle is expected but not GA.

    Recommendation for public website chatbots:

    Keep generative orchestration and grounding ON, and hide/strip citations in the web client to maintain answer quality without added complexity.

       

Under review

Thank you for your reply! To ensure a great experience for everyone, your content is awaiting approval by our Community Managers. Please check back later.

Helpful resources

Quick Links

Season of Sharing Community Challenge Launch!

Jump in, show your community spirit, and win prizes!

Kudos to our 2025 Community Spotlight Honorees

Expanding mentorship, skilling, and AI innovation

Congratulations to the May Top 10 Community Leaders!

These are the community rock stars!

Leaderboard > Copilot Studio

#1
Valantis Profile Picture

Valantis 277

#2
11manish Profile Picture

11manish 206

#3
sannavajjala87 Profile Picture

sannavajjala87 156 Super User 2026 Season 1

Last 30 days Overall leaderboard