Check out this cool CFD simulation I did with my "in house code."

[petebachant@fediscience.org]

Check out this cool CFD simulation I did with my "in house code."

Just kidding. I faked this in 5 seconds using AI. Are journals requiring source code and reproducibility checks for all submissions yet?

#openscience #cfd #aislop

[mdziemann@genomic.social]

@petebachant I never thought computational reproducibility would be necessary to distinguish real from fake research, but here we are... But what if the code and data are also fake?

[jorismeys@mstdn.social]

@mdziemann
Has been a long time. You don't need AI to fake it.
@petebachant

[petebachant@fediscience.org]

@JorisMeys @mdziemann true. Integrity checks are going to keep getting more challenging though.

[jorismeys@mstdn.social]

@petebachant In this case actually not, because something that's generated using AI looks dodgy for someone who actually knows how it's supposed to look like. It's far more difficult to distinguish a real result from one using the correct models but with generated data. And that's a battle we're fighting for a very long time already.

@mdziemann

[giuseppebilotta@fediscience.org]

@petebachant ouf, that's a very good point