ITG NowHome › AI glossary › Knowledge cutoff
What is Knowledge cutoff?
Short answer
A knowledge cutoff is the point in time after which an AI model has no training data, so it may not know newer events.
- Category Models
- Also called training cutoff, cutoff date
- Updated 20 Oct 2026
Knowledge cutoff, explained simply
A model learns from data collected up to a certain date. Anything after that is missing from its memory.
If you ask about very recent news, it may guess or say it doesn't know.
Some tools fix this by searching the web and giving the model fresh pages to read.
On ITG Now, our daily AI news stories cover what is new, which is exactly what a model with an old cutoff may miss.
Related terms
Training dataLLMHallucinationRAGChatbot
Questions people ask about Knowledge cutoff
Can an AI know about today's news?
Only if it can search the web or you paste the news in. Otherwise it relies on older training.
Why do models have a knowledge cutoff?
Training takes time, so the data must be fixed at some point before release.