Gerda 2.5 h Klausuren korrigiert - im Rotationsverfahren. Puh, wenn ich mir vorstelle, das alleine machen zu müssen ...
This year, I am really feeling the extra pain introduced by bullshit generators, aka "AI", when it comes to evaluating research assignments. I keep second guessing whether things have actually been written by certain students or not. Especially those who've talked to me about how they've been using "AI" for their other projects. Le sigh 🫠
To reduce the extra burden of checking whether the cited references actually exist or not (thanks no thanks "AI"!), I have made it mandatory for the students to add all their references to a Zotero group library. It's been so helpful! Highly recommend this
Ulrike Hahn
boosted
The latest FOSS Academic post involves more wrestling with the implications of #generativeAI for academic peer review:
https://fossacademic.tech/2025/08/06/reviewing-ai.html
In this post, I take observations from software #developers and #openSource podcasters (such as the folks at @latenightlinux ) about how genAI is swamping things like bug bounties and code reviews. This is similar to some of the issues faced by academic peer reviewers.
The latest FOSS Academic post involves more wrestling with the implications of #generativeAI for academic peer review:
https://fossacademic.tech/2025/08/06/reviewing-ai.html
In this post, I take observations from software #developers and #openSource podcasters (such as the folks at @latenightlinux ) about how genAI is swamping things like bug bounties and code reviews. This is similar to some of the issues faced by academic peer reviewers.
If you assume a 50% reproducibility rate (results from reproducibility projects vary from 12-61%), the line of irreproducible articles would be just below the "all articles" line in the logarithmic plot.
Just as a comparison to the red "paper mill products" line...
"A 2024 #researchPaper introducing Med-Gemini included the hallucination ... and nobody at #Google caught it. When Bryan Moore, a board-certified #neurologist and researcher with expertise in #AI, flagged the mistake, he tells The Verge, the company quietly edited the blog post to fix the error with no public acknowledgement — and the paper remained unchanged." https://www.theverge.com/health/718049/google-med-gemini-basilar-ganglia-paper-typo-hallucination?ICID=ref_fark
archived https://archive.ph/x8dfr#selection-1721.165-1725.127
#health #medicine #science#Gemini #diagnostics#AISlop#academicChatter