My experience with the late SO was, that the original question used to be ten years old and filled with responses that were no longer relevant to the current state.

What amazes me is that an LLM can actually make sense out of all this crappy training data and often produces meaningful answers. (maybe that's a sign that we already have AGI)