Knowledge cutoff
The date after which a model saw no training data. It knows nothing later unless you tell it or it can search.
Also written: training cutoff, cutoff date.
Training data stops at a date. Anything after it — a library version, an election, a price on this very site — is outside what the model knows, and asking anyway tends to produce a confident answer based on whatever was true before the cutoff.
Models are also unreliable about what their own cutoff is. If recency matters, give the model the current material rather than asking it to remember.