Once manuscripts are fed into AI systems for a ' blind peer review', the process can no longer be truly blind. Instead of getting thoughtful feedback from real experts, we're nowadays pushed to follow AI-generated suggestions that often miss the point, weaken our ideas, or twist our original intent. For a true peer review, we need human reviewers who can create some time to engage with our work, not machines that flatten complex thought into generic patterns.
I believe journals should be transparent about how submissions are handled insofar as AI is concerned, and how AI-based feedback is used to justify decisions to accept or reject papers.
Why are you feeding AI for review process? There are academicians available for this who can use their cognitive ability to evaluate as per your needs. machine will remain machine.
We should clarify what "blind" means in the context of peer review.
"Single blind" means: the reviewer knows the authors, but not vice versa. This has the advantage that the reviewer can judge whether the submitting group has a trustworthy reputation from the past, but biases may easily come into play, e.g. 20 years ago submissions from Asian countries had quite a hard time getting into reputable journals in part because of this.
"Double blind" means: the reviewer also doesn't know who the authors are. This switches off the aforementioned biases, but it also turns off the "reputation check" possibility.
Neither of these principles is attacked by AI. You can have the AI make a guess who the authors or reviewers are because that is quite often identifyable by their writing style or certain bullet points.
Concluding, unquestioned AI reviews of papers are bad for quality assurance in science, but the blindness of the process is not the issue.
Abdu Al-Kadi Hello sir - How are you?
The ideal of a "truly blind" peer review is increasingly difficult to maintain in the modern academic landscape, and in many cases, the veil of anonymity is semi-transparent at best. While authors' identities are formally hidden, reviewers can often make educated guesses based on citation patterns, highly specific methodological expertise, writing style, preprint server postings, or the self-citation necessary to frame one's own previous work. Furthermore, the rise of highly specialized, tight-knit research communities means there are often only a handful of global experts on a given niche topic, making anonymity a polite fiction. Consequently, while the process remains "blind" in principle and helps mitigate the most overt biases, its absolute blindness is often compromised, prompting ongoing debates about moving towards open, signed reviews to prioritize transparency over an anonymity that is frequently illusory.
I don't agree with this that only few experts are doing niche research. I do a lot of review for papers coming from different parts of the world. How would you get to know whose paper it is. This is publisher's responsibilty to send the papers without the author's name and affiliation. So far, whatever I reviewed was anonymous and I couldn't make any gusses who has written it. Either my expertise in guessing the author is poor or this statement is baseless. I hope I am wrong. If anyone else can confirm if they are able to guess whose paper they are reviewing is that easy to guess.
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress