That is what distinguishes Wikipedia from large language models or Google which'll say they have information on a given topic even if their sources are questionable. The value-add of Wikipedia is curation, and when the quality of one's outputs is a function of one's inputs, it's sometimes better to just not give an output when the input doesn't exist.
Which is, of course, completely ironic. Wikipedia's notability criterion depends heavily on books in some library stack that essentially no one will ever actually check or appearances in other print form that no one will check either.
I've found a couple of minor "folk histories" of things that appear to not have been true after looking at actual books (which may have not been actually accurate either though they seemed plausible).
(There's a WP: page about this problem that I think gives it a name, but I can't find it right now.)
Aha, someone else mentioned it in these comments: "citogenesis", and it was indeed Randall Munroe's coining, and the meta page is
https://www.wikiwand.com/en/Wikipedia:List_of_citogenesis_in...
So I feel like Wikipedia's rule don't necessarily make their content any more reliable. It just adds a step if you're looking to push some kind of biased/iffy opinion onto a page.
Wikipedia has its own "Wikipedia Library", which is a system to allow active editors access to high-quality sources. But if it has to be resorting to Sci-Hub or libgen to check sources for Wikipedia, that's also a form of public service.
Sadly that happens, and it happens a lot more than most people realize. It's way too easy to fake a reputation on Wikipedia.