Earlier quoted context omitted.
The main issue is review of material, outside of being a domain expert how would an end user determine which content to trust as rigorous?
how do i know from a DOI/name/title that the research is good? just "being" on the journal is more important than the details of peer reviewers and their own research? i am saying, how do you evaluate the "goodness" of a paper if you weren't told it came from Elsevier or some other publication? i am genuinely curious
I think I understand the motivation for your question and why you're confused.
Your mental model for your question seems to be this: if science person is competent, he/she should be able to evaluate any Libgen paper without needing to outsource the assessment of "good/bad" to a middleman publisher like Elsevier. Therefore, a file-hosting website like Libgen is all that academics should need.
But the mental model is this: the scientists/readers are busy and they don't want to waste time on bad papers. E.g. Nature journal rejects ~93% of 10000 submissions in a year.[1] Therefore, reading Nature means people don't have to wade through ~9000 other papers. In short, readers still want curation because it saves time.
Platforms like Libgen or Scihub solve distribution/downloads of pdf files. But hosting documents is the easy part. The hard problem they don't solve is the human curation. Conceivably, an academic "Libgen" would host all 10000 submitted papers and busy readers are not interested in that. Instead, having papers pass several levels of human curation filters into a manageable subset for readers leads to accumulated prestige for both the Journal and the particular paper. Prestige also leads to academic promotions, lab funding, etc. Libgen as a pdf hoster does not have a prestige-accumulation feedback loop.
(I didn't downvote your question.)
[1] https://academia.stackexchange.com/questions/8755/what-propo...
EDIT reply to: >I don't understand, how does sci-hub undermine the prestige-accumulation feedback loop?
I was not saying Scihub undermines prestige at journals like Nature. Instead, Scihub does not _solve_ the problem of creating prestige/impact for submitters (including unknown ones) who need papers validated by peers (especially respected peers). Scihub only solves the files hosting aspect -- the pdf download button. But hosting the "download links" is not the hardest problem that middlemen publishers solve.