I usually love the work of Philip Ball, but what the hell is this:

"Is AI leading to a reproducibility crisis in science?"
nature.com/articles/d41586-023

I mean the first many labs project was published in 2012! We really don't need AI - we have been perfectly capable of creating a reproducibility crisis in science all by ourselves. Humans and GlamMagz like the rag where this article is published in, are all it took. This article just completely ignores all the data we already have.

#openscience

Follow

@brembs
I am not sure if the title necessarily implies there is no reproducibility crisis yet? The point that there might be a crisis purely based on misapplied AI methods, regardless of a repro crisis for other reasons, seems to be reasonable to me

@levolz

To me it looks like it and I did not see a prominent mention of the ongoing crisis in the article, either (but I admit I didn't read it very thoroughly!).

@brembs I don't think it's in there either, but there's definitely a plethora of crises going on - so adding *this* reproducibility crisis seems fine by me

Sign in to participate in the conversation
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.