LEXICON Open App →
HomeGlossary › overfitting

overfitting

🤖 CT-AI

Official ISTQB Definition

A modeling error that occurs when a machine learning model learns the training data too well, including noise, resulting in poor performance on new data.

3 Ways to Think About It

💡
The Quick Take

When an AI memorizes training data instead of learning general patterns - it passes training tests but fails in production.

🔍
Look Closer

The #1 enemy of ML models: performing perfectly on test data but badly on real-world data.

🎯
The Bottom Line

An AI that's too specialized to its training examples, caught by testing on held-out data it's never seen.

Practice this term with quizzes and arcade games

Study with Lexicon →
← organizational test strategyAll Termspair programming →