Home/Glossary/Knowledge Cutoff

AI Glossary

Knowledge Cutoff

The date after which an AI model has no information

Definition

A knowledge cutoff (or training cutoff) is the date at which an AI model's training data ends. The model has no knowledge of events, publications, or changes that occurred after this date. This is why a model with a 2024 cutoff doesn't know about products, news, or research released in 2025. Retrieval tools and RAG can supplement models with current information.

Related Terms

Back to Glossary