Skip to main content
Home/Risks/IBM2025/Improper retraining

Improper retraining

Sub-category
Risk Domain

AI systems that fail to perform reliably or effectively under varying conditions, exposing them to errors and failures that can have significant consequences, especially in critical applications or areas that require moral reasoning.

"Using undesirable output (for example, inaccurate, inappropriate, and user content) for retraining purposes can result in unexpected model behavior."

Supporting Evidence (1)

1.
"Repurposing generated output for retraining a model without implementing proper human vetting increases the chances of undesirable outputs to be incorporated into the training or tuning data of the model. In turn, this model can generate even more undesirable output."

Other risks from IBM2025 (63)