AI Judge changed title after evaluation, original title: "Thomas G. Dietterich reminds arXiv authors of content responsibility"
The notice prompted researcher discussion on enforcement steps such as temporary bans for issues like hallucinated references following recent uploads.
Attention @arxiv authors: Our Code of Conduct states that by signing your name as an author of a paper, each author takes full responsibility for all its contents, irrespective of how the contents were generated. 1/
Many users criticized arXiv's author responsibility rules for AI-generated papers as enabling pollution of science or overly harsh enforcement, while others supported them for ensuring accountability and blocking low-quality submissions.
No Digg Deeper questions have been answered for this story yet.
@JimDMiller @tdietterich @arxiv “Beyond the ability” is quite simply laughable. If they aren’t able to even verify the citation about the subject they should not be publishing papers about the same subject.
Making humans responsible for their AI use seems like an incredibly reasonable way to address problems & opportunities in the use of AI for academic research, at least in the short term (autonomous scientific work will require different solutions).
Attention @arxiv authors: Our Code of Conduct states that by signing your name as an author of a paper, each author takes full responsibility for all its contents, irrespective of how the contents were generated. 1/
Attention @arxiv authors: Our Code of Conduct states that by signing your name as an author of a paper, each author takes full responsibility for all its contents, irrespective of how the contents were generated. 1/
What if the citation is to a source that is not available online and is in a language one of the authors doesn't understand? What if the paper has a hundred citations and one of the authors is an extraordinarily productive scholar who has better things to do than check the accuracy of citations that other authors put into the paper?
@JimDMiller @predict_addict @tdietterich @arxiv Can't take responsibility, don't take authorship. Simple.
@JimDMiller @CCguerilla @predict_addict @tdietterich @arxiv @JimDMiller , basic research ethics is that you cite things you've read. If it's in a different language, get it translated.
@JimDMiller @tdietterich @arxiv You are a disgrace to your field and academics everywhere. How such an unserious person got to be employed by an institution as prestigious as the University of Chicago is completely beyond me.
@JimDMiller @CCguerilla @predict_addict @tdietterich @arxiv No, but one of the authors needs to be responsible for it. And if you can't trust your coauthors to vet citations, you shouldn't work with them. Or, you just be an adult and do it yourself. Do you check your citations? If so, what's your issue with it?
The penalty is a 1-year ban from arXiv followed by the requirement that subsequent arXiv submissions must first be accepted at a reputable peer-reviewed venue. 4/
Examples of incontrovertible evidence: hallucinated references, meta-comments from the LLM ("here is a 200 word summary; would you like me to make any changes?"; "the data in this table is illustrative, fill it in with the real numbers from your experiments") end/
@JimDMiller @CCguerilla @predict_addict @tdietterich @arxiv Sure, but this is a price to pay for academic integrity. If you can't trust your students, then you should do it yourself.
@JimDMiller @tdietterich @arxiv Fucking yes. This is a research paper, not an entertainment product. What do you think the citations are there for?
@tdietterich @arxiv Very good, please strongly enforce it.
Attention @arxiv authors: Our Code of Conduct states that by signing your name as an author of a paper, each author takes full responsibility for all its contents, irrespective of how the contents were generated. 1/
We must think about how to handle the fact that LLMs can generate papers without any human intervention, but this is not the way. Also, I want to see the arXiv ban senior PIs who upload 40 papers a year.
Attention @arxiv authors: Our Code of Conduct states that by signing your name as an author of a paper, each author takes full responsibility for all its contents, irrespective of how the contents were generated. 1/
How do you tell whether authors agreed to be listed? If person X writes an AI generated slop paper, puts my name on it without my knowledge, and uploads it to the arxiv, am I banned too? Conversely, if person X and I write slop AI and he submits it to the arxiv and gets banned, can I then feign ignorance of the project and avoid punishment? (I don’t envy your job, and think you’re doing great, in difficult and complicated times!…)
@JustinAngel @arxiv I agree that there could be biases in our pipeline. We apply a standard LLM detection algorithm to identify papers that need scrutiny. Moderators may also be biased. We would love to collaborate with researchers to study the bias and effectiveness of our operations!
A few more thoughts: I agree with the concern that Arxiv will become full of millions of AI-slop papers written by agents. **But** this is not the answer. Not only because banning an author for life is not fair (it might that you write only some sections and ofcourse that you can't check all the references one by one), but also because this is not the real problem. AI is writing our papers at this point, and the only question is at what level. If I ask the model to rephrase a sentence and it keeps, " Sure, I will do it for you", it doesn't mean anything about the quality of the paper. The same as we didn't punish papers with grammar errors in the past. Moreover, I can create entire fake papers that will pass this bar from one prompt, and soon, we will have systems that will catch these mistakes automatically. At the end, we need to come up with a semi-automatic system that will check for correctness without the question of whether it was generated by AI (the question of whether we want humans to be responsible for each paper is a different story)
We must think about how to handle the fact that LLMs can generate papers without any human intervention, but this is not the way. Also, I want to see the arXiv ban senior PIs who upload 40 papers a year.
@tdietterich @arxiv My guess is that this policy will be applied selectively depending on institutional privilege and personal notoriety. It'll end up as a tool of silencing unconnected individuals vs. promoting better scientific discourse.
I aspire to be wrong.
@JimDMiller @tdietterich @arxiv ArXiv is primarily computer science preprint website, not some ancient mythology site or a site about Egyptian stone scribbles.
Every meta-analysis and review on LLM writing detection says the technology doesn't work. How will you apply a biased technology in a fair and consistent way?
"human detection accuracy varied widely but generally clustered around chance performance" (Ramos, 2026) @ https://arxiv.org/pdf/2604.03437
