AI governance reviewLucenOne working template · adapt to the applicable framework
lucen.onePurpose and sources
Specific purpose of the function
Authorized sources
Excluded sources
Delegation
Task that may be automated
Task that may be assisted
Task that may be suggested
Decision that cannot be delegated
Control
- Are sources visible?
- Are uncertainties explicit?
- Can the user correct or reject the output?
- Is meaningful human action traceable?
- Is there a withdrawal mechanism if the function fails?