Nobody is unaware of the knowledge cutoff, and sharing the Wikipedia article is not helping anyone. Your point is easily rebutted by taking whatever open weight

Nobody is unaware of the knowledge cutoff, and sharing the Wikipedia article is not helping anyone. Your point is easily rebutted by taking whatever open weights/source model has an outdated cutoff and training or fine tuning it on more data, which is again always going to be viable given a modicum of compute