Yes the title should be "Could machine learning fuel yet another reproducibility crisis in science?" There's a bit of denial about the framing of the article,

Yes the title should be "Could machine learning fuel yet another reproducibility crisis in science?" There's a bit of denial about the framing of the article, which is maybe telling. The article is about a slightly different form of reproducibility crisis I think, one more specific to ML studies. But the title maybe isn't quite right.