← Back to the AI Glossary 📖 Data & Evaluation

Data Drift

When the real-world data an AI system encounters in production gradually shifts away from the data it was originally trained or evaluated on, degrading performance over time.

This is why AI systems in production usually need ongoing monitoring — a model that worked well at launch can quietly become less accurate as the world it's operating in changes.
One of 60 free AI glossary terms
Plain-language definitions for the AI jargon you'll actually run into — no email needed, ever, for this section.
Browse the Full Glossary →
More in Data & Evaluation
Eval (Evaluation)
A structured test set up to measure how well an AI model or AI-powered feature performs at...
Leaderboard
A public ranking that compares AI models against each other on one or more benchmarks, oft...
Ground Truth
The verified, correct answer used as a reference point to check whether an AI's output is...
Benchmark
A standardised test or dataset used to measure and compare how well different AI models pe...