According to Guillame Cabanac, posting on PubPeer, this editorial1 previously contained the extraneous phrase “Regenerate response”. This phrase has been removed, and the correction statement has been added:
This article has been republished with minor changes. These changes do not impact the academic content of the article.1
While the authors have openly declared the fact that they have made a correction, the nature of the correction is unspecified. In my opinion, therefore, this still qualifies as a stealth edit.
The purpose of formal errata and corrigenda is transparency. The reader should be informed as to the precise nature of the error so that they may come to their own conclusion rather than taking the authors’ word that the changes “do not impact the academic content.” Perhaps they do not, but I would argue that the apparent use of an AI system in drafting the article absolutely does have implications for its content.
The Taylor & Francis editorial policies appear to agree, stating the following:
Where AI tools are used in content generation, they must be acknowledged and documented appropriately in the authored work.
This policy applies to all Taylor & Francis Group journals. Since no such acknowledgement or documentation appears in the editorial by Velthuizen and Otto, if an AI tool was used, the article is in violation of this policy. Needless to say, this is a poor example for the editors of the journal to set.