Underfitting
A modelling error where a machine learning model is too simple to capture the underlying patterns in the data, resulting in poor performance on both training and test data.
In plain language
When an AI is too simple to learn the patterns in the data. Like trying to draw a curve with a straight line; it misses important patterns and performs poorly on everything.
Why this matters
Underfitting is a model quality risk that governance frameworks should address through validation testing. AI systems that underfit fail to deliver business value and waste implementation investment. Your testing processes should verify that model complexity is sufficient for the task.
Relevance
ImplementationIdentifies a failure mode in model development that standard metrics can overlook, requiring governance oversight of model complexity and validation processes.
Related terms
Putting underfitting into practice in your organisation?
Ready to transform your AI strategy?
Partner with Australia's AI strategy and governance specialists. From adoption roadmaps to ISO 42001 audit readiness.