Knowledge Cutoff
The date when an AI model’s training data ended — everything after this is unknown to the model unless retrieved at query time.
Plain Language Definition
A knowledge cutoff is the specific date when an AI model stopped reading new web pages and books during its training. After that date, the model’s built-in knowledge is frozen — it does not know about events, products, or companies that emerged afterward unless retrieval augmentation (live web search) is active. This is why AI models sometimes give outdated information.
Technical Definition
Temporal boundary of an LLM’s offline parametric training dataset, rendering the model’s internal weights blind to events, publications, and content published after the cutoff date — addressed at inference time only by retrieval augmentation over live indices.
Why This Matters for AI Search Visibility
If your company launched or significantly evolved after a model’s training cutoff, the model’s base knowledge about you may be nonexistent or outdated. This makes real-time retrieval optimization (GEO and RAG-focused strategies) more important than LLMO for newer businesses.