-
Suggestion
-
Resolution: Unresolved
-
None
Problem
Currently, the AI definition feature in Confluence allows any user to edit AI definitions, regardless of their editing permissions within a space.
- This poses a risk of inaccurate information being introduced, as control over editing these definitions is not aligned with the existing space-level permissions.
- While users can turn off this feature individually, there is no option to disable or restrict this feature at the space level.
Suggested Solution
Implement a feature that allows space administrators to control the ability to edit AI definitions:
- This could include options to completely disable AI definitions within a space or restrict editing capabilities based on user permissions in that specific space.
Why This Is Important
The integrity and accuracy of information within a Confluence space are critical for teams relying on shared knowledge and documentation.
- Allowing unrestricted edits to AI definitions can lead to misinformation and reduce trust in the content.
- Granular control over this feature would enhance alignment with existing permission structures and help maintain accurate and reliable documentation.
Workaround
Currently, users can individually turn off definition highlighting for themselves by following the steps detailed in the Confluence documentation.
However, this is not a comprehensive solution as it does not prevent others from editing or viewing AI definitions.
Note:
The feature also provides the edit history to show transparency and traceability in the definition changelog, for example, if the new definition is visible to anyone viewing the page, space or the entire site.
- is related to
-
AI-110 Ability to disable/enable Atlassian Intelligence (AI) for only a specific space/project (not the entire Confluence / Jira / etc. site)
- Gathering Interest
-
AI-903 Ability to enable Atlassian Intelligence at a project level
- Gathering Interest
-
ROVO-107 Enhanced Granular Control Over AI Features
- Gathering Interest
[ROVO-11] Limit AI Definition editing permissions at space level in Confluence
Comment |
[ Holy smokes...this one is particularly bad.
So we can't stop anyone from editing definitions, as we could normally restrict people from so many other action, but *we cannot opt ourselves out of Rovo or get Atlassian support to remove it from our sites???* ] |
Description |
Original:
h3. Problem
Currently, the AI definition feature in Confluence [allows any user to edit AI definitions|https://intercom.help/rovo/en/articles/9479386-knowledge-cards-and-definitions#h_17e11ede90], regardless of their editing permissions within a space. - This poses a risk of inaccurate information being introduced, as control over editing these definitions is not aligned with the existing space-level permissions. - While users [can turn off this feature individually|https://intercom.help/rovo/en/articles/9479386-knowledge-cards-and-definitions#h_1b95aa9a7b], there is no option to disable or restrict this feature at the space level. h3. Suggested Solution Implement a feature that allows space administrators to control the ability to edit AI definitions: * This could include options to completely disable AI definitions within a space or restrict editing capabilities based on user permissions in that specific space. h3. Why This Is Important The integrity and accuracy of information within a Confluence space are critical for teams relying on shared knowledge and documentation. * Allowing unrestricted edits to AI definitions can lead to misinformation and reduce trust in the content. * [Granular control over this feature|https://jira.atlassian.com/browse/AI-950] would enhance alignment with existing permission structures and help maintain accurate and reliable documentation. h3. Workaround Currently, users can individually turn off definition highlighting for themselves by following the steps detailed in the Confluence documentation. However, this is not a comprehensive solution as it does not prevent others from editing or viewing AI definitions. |
New:
h3. Problem
Currently, the AI definition feature in Confluence [allows any user to edit AI definitions|https://intercom.help/rovo/en/articles/9479386-knowledge-cards-and-definitions#h_17e11ede90], regardless of their editing permissions within a space. - This poses a risk of inaccurate information being introduced, as control over editing these definitions is not aligned with the existing space-level permissions. - While users [can turn off this feature individually|https://intercom.help/rovo/en/articles/9479386-knowledge-cards-and-definitions#h_1b95aa9a7b], there is no option to disable or restrict this feature at the space level. h3. Suggested Solution Implement a feature that allows space administrators to control the ability to edit AI definitions: * This could include options to completely disable AI definitions within a space or restrict editing capabilities based on user permissions in that specific space. h3. Why This Is Important The integrity and accuracy of information within a Confluence space are critical for teams relying on shared knowledge and documentation. * Allowing unrestricted edits to AI definitions can lead to misinformation and reduce trust in the content. * [Granular control over this feature|https://jira.atlassian.com/browse/AI-950] would enhance alignment with existing permission structures and help maintain accurate and reliable documentation. h3. Workaround Currently, users can individually turn off definition highlighting for themselves by following the steps detailed in the Confluence documentation. However, this is not a comprehensive solution as it does not prevent others from editing or viewing AI definitions. h4. (i) Note: The feature also provides the *edit history* to show transparency and traceability in the {*}definition changelog{*}, for example, if the new definition is visible to anyone viewing the page, space or the entire site. |
Affected Products | New: Confluence [ 18440 ] | |
Component/s | Original: Confluence AI [ 74210 ] | |
Component/s | New: Contextual Reading Aids [ 75094 ] | |
Key | Original: CONFCLOUD-79270 | New: AI-957 |
QA Demo Status | Original: Not Done [ 14330 ] | |
QA Kickoff Status | Original: Not Done [ 14234 ] | |
Project | Original: Confluence Cloud [ 18513 ] | New: Atlassian Intelligence [ 23110 ] |
Holy smokes...this one is particularly bad.
So we can't stop anyone from editing definitions, as we could normally restrict people from so many other action, but we cannot opt ourselves out of Rovo or get Atlassian support to remove it from our sites???