You're almost there! Please answer a few more questions for access to the Applications content. Complete registration
Interested in joining? Complete your registration by providing Areas of Interest here. Register

AI Configurator Prompt Tuning – Prioritizing Manager Input in Performance Evaluations

edited Apr 29, 2026 2:41PM in AI for Fusion Applications 1 comment

Summary:

Hi everyone,

We are currently testing AI Assist for Performance Evaluations in Oracle HCM Cloud and have observed a consistent issue.

In cases where the manager rating is “Does Not Meet Expectations” the AI-generated manager comments are still overly positive. It appears that AI Assist is pulling heavily from the employee’s self-assessment (Goals and Overall Summary comments), which is influencing the tone and making it inconsistent with the manager’s rating.

Content (please ensure you mask any confidential information):

Has anyone successfully adjusted the prompt logic to ensure that:

  • Manager rating drives the tone of the output
  • Lower ratings (e.g., Does Not Meet Expectations) produce appropriately constructive/critical summaries
Tagged:

Howdy, Stranger!

Log In

To view full details, sign in.

Register

Don't have an account? Click here to get started!