Data curation risk for AI
Description
When training or tuning data is improperly collected or prepared, the result can be a misalignment of a model's desired values or intent and the actual outcome.
Why is data curation a concern for foundation models?
Improper data curation can adversely affect how a model is trained, resulting in a model that does not behave in accordance with the intended values. Correcting problems after the model is trained and deployed might be insufficient for guaranteeing proper behavior. Improper model behavior can result in business entities facing legal consequences or reputational harms.
Parent topic: AI risk atlas