Upgrade StorageNo plugin like chatpdf in zotero?
I know zotero-gpt, papers-gpt. But I mean something like chatpdf、pdf.ai、chatdoc: once upload, hightlight some lines during reading , ask questions and chat. （zotero-gpt even doesn't wake in pdf-read window.）
I only find askyourpdf, which offer a zotero plugin, but cannot work in my system(zotero 7).
I only find askyourpdf, which offer a zotero plugin, but cannot work in my system(zotero 7).
I have a basic question: Is it okay to analyze scientific publications with AI tools? If a user uses copyrighted texts or excerpts thereof in prompts and processes them further, this constitutes a copyright infringement. The following applies: Under German law, almost every published text (including blog posts, etc.) is protected by copyright, unless the term of protection has expired (70 years after the author's death) or subsequent use has been explicitly permitted.
I see no problem when reading and annotating documents with the Zotero reader. What is your opinion on this?
- AI companies used entire repositories of pirated copyright books to train their AI:
https://arstechnica.com/tech-policy/2025/02/meta-torrented-over-81-7tb-of-pirated-books-to-train-ai-authors-say/
https://www.theatlantic.com/technology/archive/2025/03/libgen-meta-openai/682093/
- academic researchers' published journal papers have already been sold to AI companies by academic publishers, without consultation with those authors:
https://www.insidehighered.com/news/faculty-issues/research/2024/07/29/taylor-francis-ai-deal-sets-worrying-precedent
The issues may still be decided in the courts. The AI companies are claiming "fair use" as the basis for their harvesting of copyrighted material (with an additional dose of "the ends justify the means").
https://theconversation.com/meta-allegedly-used-pirated-books-to-train-ai-australian-authors-have-objected-but-us-courts-may-decide-if-this-is-fair-use-253105
https://www.nbcnews.com/tech/tech-news/openai-gets-partial-win-authors-copyright-lawsuit-rcna138705
At the same time, academic publishers are developing policies for the use of AI by academics in academic writing ...
https://newsroom.taylorandfrancisgroup.com/taylor-francis-clarifies-the-responsible-use-of-ai-tools-in-academic-content-creation/
https://taylorandfrancis.com/our-policies/ai-policy/
eg "Generative AI tools must not be listed as an author", and "Authors must clearly acknowledge within the article or book any use of Generative AI tools".
Also perhaps relevant to the potential use of AI tools within Zotero is the erosion of critical thinking skills in association with AI use, already being studied...
https://phys.org/news/2025-01-ai-linked-eroding-critical-skills.html
I am undecided because, on the one hand, I see the desire for AI tools among students. On the other hand, I am also a representative of a public institution.
In addition, the use of AI is one thing: As you say, there are university guidelines for this. Another thing is copyright: uploading documents to ChatPDF & Co is not okay, even if it is not prosecuted / provable.
If I were a student, I wouldn't care about copyright. As a digital immigrant, I grew up with "Napster", "Kino.to" etc. This discrepancy - reality VS idealism - bothers me.
Maybe it's also a cultural thing: In Germany, we plan everything perfectly first - and if there's only one catch, then we don't do it: Don't make any mistakes; "Woe betide anyone who complains!".
In America, you seem (!) to try things out first. If there are problems, they can be solved... Citavi version 7 has an AI function, and when we expressed concerns about copyright, they looked at us funny at first :-)
PS: "Die Sache ist durch", "Die Sache ist gegessen", "Die Messe ist gelesen", "Der Drops ist gelutscht", "Das Kind ist in den Brunnen gefallen" -
We have many expressions for your metaphor.
Have a nice day!
If openAI then uses those files illegally, that's their problem -- but there are of course many services that don't use chats/uploads for training (Claude, chatGPT pro, let alone local LLMs)
i find your hint that we as users can make a choice by dealing with the tools particularly productive.
i will never be able to set up a local LLM. but Claude f.e. sounds good.
thank you too, @adamsmith, for your assessment and tips!