2

Can I improve a paper peer-review using ChatGPT?

The goal is not to generate content. This is to improve my review of a conference paper in terms of harmful content moderation (things that authors might find offensive) and the overall narrative.

Let's say that I write a review such as:


The paper does A, B, and C.

  • It does not evaluate A in X methodology
  • The B section has strong assumptions.
  • C is a good experiment.

The paper is a good contribution ready for publication. The above changes will make it stronger publications.


Then I run on ChatGPT, read the answer, to modifications on the response and submit the answer.

Is this ethically okay according to reviewing principles?

From my side, I often get very bad reviews that are hard to review, and more often that I will like to assume I find them offensive. So personally, at this stage, I wouldn't mind if someone provides me a review following the above methodology.

1
  • 4
    To be honest, I much prefer receiving reviews that consists of a list of concise, clear single topic comments i can then work through when formulating a reply, than getting a long rambling text, where I have to extract the same points first, in the hope of not missing anything.
    – mlk
    Apr 3 at 13:31

1 Answer 1

5

Let's avoid the discussion about the content/quality (of Chatwhatever AI) and let's stick to the formal part.

Probably the worst thing you are doing is uploading some text that should be confidential (your review) to a service over the web on which you have no control nor you know where/how it is stored.

3
  • Very good point. But I dont upload the paper. Only my review, which is my own content. Does this issue still persist? Apr 3 at 9:45
  • 1
    @CarlosMougan I explicitly wrote "your review". Yes, you review is confidential and that is the problem.
    – EarlGrey
    Apr 3 at 10:25
  • Many thanks! I see your point! Apr 3 at 10:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .