Public AI Constitution project describes a framework for AI behavior
The project says explicit priorities can help a model navigate conflicts between honesty, helpfulness and harm avoidance. Its website describes the constitution as open for community review.
TLDR
The Public AI Constitution Project says its constitution makes intended model values and behavior explicit, giving training a definite target. It argues that a hierarchy of priorities can help resolve competing demands, such as being helpful while avoiding harm. Beyond guiding behavior, the project says the constitution can give developers, users and regulators a shared reference for explaining problems after they occur.
Combined views
—
1 Source, first seen ago
