Back

AI and data privacy in Performance

Each AI feature in Performance uses only the data needed for its task and shows its results only to people allowed to see them. This article explains, feature by feature, what data is used, who sees the result and what admins control. It's useful when answering questions from employees, managers, works councils or data protection officers.


 

Before you start

  • Admins control who can use One in Settings > Permissions > One.
  • The visibility settings of each review, such as Restrict results visibility and Grant managers early access to responses, also control who sees AI Insights.


 

Steps

Check what each AI feature uses and who sees the result

Feature Data it uses Who sees the result Who controls it
AI Insights The questions and answers of one reviewee's published evaluations in that review Admins with full visibility, review owners, the reviewee's managers (timing follows the review's visibility settings) Review visibility settings
Team insights The AI Insights of the people in a manager's team Admins, review owners, the manager and higher managers Review visibility settings
Enhance Only the text in the editor you're writing in Only you, until you accept and submit Use One permission
Suggested competencies The job catalog node and the Factorial competency catalog Admins with Manage competencies Manage competencies permission
One chat The data the user can already see, fetched when they ask Only the user who asked Use One • each area's access
One Analytics Company performance data Admins only Admin role


 

Limit who can use AI with One

  1. Go to Settings > Permissions > One.
  2. Choose who has Use One: specific roles, specific employees, admins only, or everyone.
  3. Save.


 

Rules and behavior

  • If someone can't see a review's results in Factorial, then they can't see its AI Insights and One can't show them its data.
  • If Restrict results visibility is on for a review, then results, including AI Insights, stay limited as that setting defines.
  • If an AI feature produces text about a person (AI Insights, Team insights), then it's a summary of reviewers' answers. It doesn't change scores.
  • If One is asked to give or delete feedback, then the user must approve before anything is saved, and the action is recorded as theirs.
  • If you remove Use One from someone, then they no longer have One chat, Enhance or One Analytics.


 

Limitations

  • There's no company setting to turn off AI Insights or Team insights.
  • Enhance and One can be limited only through the Use One permission, not per module.
  • Nobody can edit AI Insights to correct them. Correct the underlying answers instead, which regenerates the insight.


 

Troubleshooting

  • Symptom: "A manager saw AI Insights before the review ended." → Cause: Grant managers early access to responses is on for that review. → Fix: Turn it off in the review's visibility settings for future reviews.
  • Symptom: "Someone who shouldn't use One still can." → Cause: Their role or name is included in Settings > Permissions > One. → Fix: Remove them there.


 

FAQ

1. Does One show my review answers to colleagues?

No. One only shows each person the data they can already see in Factorial.

2. Can we turn off AI in Performance?

You can remove Use One to turn off One, Enhance and One Analytics for a group. AI Insights can't be turned off.

3. Is our data used to train AI models?

Read Factorial's AI and data processing terms, or contact your account manager.

4. Does AI make decisions about employees?

No. AI summarises and suggests; people make every decision, and every change One makes needs approval.

Was this article helpful?

Give feedback about this article

Can’t find what you’re looking for?

Our customer care team is here for you.

Contact us

Knowledge Base Software powered by Helpjuice