Retraction crisis?

[Frabjous]

I was reading this article
https://www.technologyreview.com/2024/05/15/1092535/a-wave-of-retractions-is-shaking-physics/
Are the numbers actually up? I am not saying that processes cannot be improved.

 

[Vanadium 50]

Its hard to tell if the numbers are up because it is, by definition, hard to count the ones that don't get caught. Further, there is no consensus as to what is retractable: fraud, certainly. But what about unreproduced results? Do we retract Cabrera's monopole paper because he saw only one, and nobody else saw any? Do we retract Milikan's oil drop experiment because he was subjective about what data to keep? Do we retract Michalson-Morely because their error analysis isn't up to modern standards?

Fraud is not new. I am old enough to remember Element 118 when Victor Ninov managed to bamboozle not just the community but his own experiment.

I do suspect there is more:
1. There are more papers, so there are more fraudulent papers.

2. China. There is a weird synergy going on, when China wants its researchers to publish in Western journals, and Western journals want to show how international and multicultural they are. But whatever the CCP's motivation, advancing science is not it.

3. AI makes it easier than in the past to forge data.