Like there’s thousands of “machine learning” papers published in MDPI and Hindawi journals that are literally just a writeup of how the authors applied scikit-learn algorithms to a small dataset. No conclusions or advances in understanding are made.
Consequently these are extremely easy to filter out as trash. I absolutely don’t condone or dismiss the problem, but its not yet a serious impediment to real research. It’s mostly self contained within this ecosystem of trash journals. Trash research doesn’t escape containment to contaminate more serious venues because it’s transparently and obviously trash and serious journals care about their reputation.
So, I think its important not to overstate the consequences or reach of this problem. That’s not to say academic publishing doesn’t have problems, but I don’t think this kind of trash publishing is going to seriously impede progress. That said I would shut down all these people and their journals if I could