ChatGPT — Conflicts of APCs

Even a new AI agent understands the conflicts of interest around APCs

A friend asked ChatGPT to list all the editorial conflicts of interest associated with scholarly journals charging authors an article processing charge (APC). While missing some of the more subtle sources of corruption introduced by APCs — wooing top authors with waivers or discounts, or creating discounts for multiple submissions to drive quantity harder, for instance — the response is quite robust:

There are several editorial conflicts of interest that can arise when scholarly journals charge authors an article processing charge (APC):

Bias in peer review: If a journal’s financial stability depends on APCs, it may be more likely to accept articles that are subpar in quality in order to maintain its revenue stream. This could compromise the quality of peer review, as reviewers may feel pressure to approve articles in order to keep the journal financially viable.

Conflicts with peer reviewers: Peer reviewers who are paid for their work may feel conflicted about their role as impartial evaluators of research if they are also aware that the journal’s financial stability depends on APCs.

Conflicts with editors: If editors are aware that the journal’s financial stability depends on APCs, they may be more likely to accept articles that are of lower quality in order to maintain revenue. This could compromise the overall quality of the journal and damage its reputation.

Bias in selection of articles: If a journal’s financial stability depends on APCs, it may be more likely to prioritize articles that are more likely to generate high APC revenue over articles that are of higher quality or more relevant to the journal's focus.

Conflicts with authors: If authors are aware that the journal’s financial stability depends on APCs, they may feel pressure to pay the APC even if they are dissatisfied with the quality of the peer review or editing process. This could create a perception that the journal's quality and impartiality are for sale.

The conflicts with reviewers may exist whether the reviewer in question is paid — for example, an Associate Editor receiving a small stipend to run a section of a journal — or not.

APCs introduced conflicts of interest — that is, authors paying editors — which seemed unacceptable. However, we’re well past that point.

So, given our apparent ability to self-corrupt, maybe the age of the AI overlords will actually make more sense than the age of self-serving techno-utopianism and scientific careerism.