By default, NULL values are not treated as distinct entries. Specifying NULLS NOT DISTINCT on unique indexes / constraints will cause NULL values to be treated distinctly.
Specifies whether for a unique index, null values should be considered distinct (not equal). The default is that they are distinct, so that a unique index could contain multiple null values in a column.
For the purpose of a unique constraint, null values are not considered equal, unless NULLS NOT DISTINCT is specified.
I think saying that NULLs are distinct by default is the better way to present this. The documentation disagrees on the verbiage but both state the same truth; the feature matrix page needs to be fixed.