No open-access location is known for this record yet.
No open-access location is known for this record yet.
The aim of the European Heart Journal (EHJ) is to publish innovative, methodologically sound, and clinically relevant research manuscripts that can change clinical practice and substantially advance knowledge on cardiovascular diseases.1–5 The EHJ is committed to publishing only the best cardiovascular science adhering to the highest ethical principles.1–5 The EHJ uses highly rigorous peer review, critical statistical assessment, and highest quality editorial process, to ensure the novelty, accuracy, quality, and relevance of all accepted manuscripts in order to inspire clinical practice and reduce the global burden of cardiovascular diseases.1–10 In addition, the EHJ pays special attention to improving diversity and inclusion at all stages in the publishing process.1–2 The journal also responds to new initiatives from major funders of public research, e.g. by offering authors the option to publish open access. These items have been addressed at length in a recent special article about the journal’s quality standards.10 The EHJ subscribes to the recommendations of the International Committee of Medical Journal Editors (ICMJE) for the conduct, reporting, editing, and publication of scholarly work in biomedical journals.5 These recommendations include criteria for authorship (i.e. the ‘4 criteria’, designed to guarantee that all authors accept full responsibility and stand for the integrity of the entire work), disclosure of potential conflicts of interest (using the electronic ICMJE uniform standardized disclosure form), trial registration (in selected, high-quality, publicly accessible, open repositories that include the 24-pre-defined registration items), data sharing policies (emphasizing the importance of detailing the data sharing plan), and more recently, defining the editorial requirements for pre-prints (original version of the work, directly posted by the authors on not-for-profit pre-print servers, before undergoing peer review).4–10 The ICMJE continuously updates its recommendations to reflect changes in the research community. In May 2023, ICMJE announced some important updates to its recommendations, and there was a particular focus on the role of artificial intelligence (AI) in academic publishing.5 Most changes in the May 2023 update to the ICMJE recommendations addressed the use of AI in the scientific and editorial process.5 The section entitled ‘Roles and Responsibilities of Authors, Contributors, Reviewers, Editors, Publishers, and Owners’ has been modified to highlight issues related to AI.5 When defining the role of authors and contributors, a new section has been added to explicitly address the role of AI, entitled ‘Artificial Intelligence (AI)-Assisted Technology’. In the new section the ICMJE states the following: ‘At submission, the journal should require authors to disclose whether they used artificial intelligence (AI)-assisted technologies (such as Large Language Models [LLMs], chatbots, or image creators) in the production of submitted work. Authors who use such technology should describe, in both the cover letter and the submitted work, how they used it. Chatbots (such as ChatGPT) should not be listed as authors because they cannot be responsible for the accuracy, integrity, and originality of the work, and these responsibilities are required for authorship…5’ The EHJ follows the ICMJE recommendations and will introduce the following requirements for authors submitting to the journal: At the point of submission, authors must declare if AI or AI-assisted technologies were used in the preparation of the manuscript. Authors should also provide this declaration in the cover letter. In a separate section in the manuscript (which could be the methods section), authors must clearly describe how AI or AI-assisted technologies were used in the study and/or preparation of the manuscript. The use of AI or AI-assisted technology could include the use of LLMs, chatbots (such as ChatGPT), and image creators. Artificial intelligence or AI-assisted technologies do not qualify as authors and must not be listed authors or co-authors on a manuscript. Similarly, they should not be listed as authors or co-authors in references. Authors must declare and confirm that no part of their submission, which includes any content arising from AI or AI-assisted technology, is under consideration or has been submitted to another journal and that no element of their submission, including any content arising from AI or AI-assisted technology, constitutes plagiarism or breaches the journal’s ethical policies. The ICMJE recommendations also remind authors to pay close attention to the output of AI or AI-assisted technologies, particularly chatbots such as ChatGPT. While the results of AI tools may sound convincing and authoritative, AI tools cannot distinguish between true and false information, meaning that their output could be incorrect or misleading, or provide an incomplete or biased result.5 Humans are responsible for any submitted material (including that generated by AI technologies) and must ensure appropriate attribution including full citations. Finally, ICMJE updated the ‘Manuscript Preparation and Submission’ section to state that ‘referencing AI-generated material as the primary source is not acceptable’.5 The EHJ agrees with this principle. The ICMJE recommendations relating to peer review have also been updated to reflect developments in AI. First, reviewers should not upload the manuscript to software or other AI technologies where confidentiality cannot be assured.5 Second, reviewers should disclose to journals if and how AI technology was used in the process of evaluating the manuscript or in writing their reviewer comments. As noted previously, ICMJE warns that the output of AI or AI-assisted technologies can be incorrect, incomplete, or biased.5 The World Association of Medical Editors (WAME) also has given the following recommendations on ChatGPT and chatbots in relation to scholarly publications11: Chatbots cannot be authors. Authors should be transparent when chatbots are used and provide information about how they were used. Authors are responsible for the work performed by a chatbot in their paper (including the accuracy of what is presented and the absence of plagiarism) and for acknowledging all sources (including material produced by the chatbot). Editors need appropriate tools to help them detect content generated or altered by AI, and these tools must be available regardless of their ability to pay. The new ICMJE recommendations also address other issues. It is now clarified that all authors must have ‘critically reviewed’ the work for important intellectual content and must have the opportunity to make changes.5 Authors are also encouraged to use a more inclusive language and follow the Sex And Gender Equity in Research (SAGER) guidelines for reporting sex and gender information.5 Artificial intelligence technologies will soon completely revolutionize clinical practice as we understand it today.12–18 The AI improves diagnosis and clinical decision-making and therefore, should help to improve patient outcomes. Indeed, AI should not be considered a threat but rather as a powerful tool to help clinicians and patients. Similarly, it is also evident that AI technologies are inducing a major paradigm shift in the scientific and editorial process. This represents a unique opportunity. For instance, some AI tools such as ChatGPT can provide helpful reference lists related to a specific topic or summarize a complex idea in accessible terms. These technologies use generative language models to ‘generate’ content based on large datasets that are fed into their systems. They may write reviews and their output can be coherent, logical, and even natural. But, as with all disruptive innovations, they also carry unexpected negative consequences. The AI is trained from existing literature that may be inaccurate, out of date, or not representative. Another problem is the ‘dataset shift’ that occurs when there is mismatch in contents between the dataset used to develop the system and the dataset where the system is ultimately deployed.12–18 The AI system may also suffer from ‘hallucinations’, namely confident responses by an AI system that are not justified by its training data. Currently, this is considered a major problem in some LLMs.15–17 Finally, of major concern, ChatGPT is not transparent and may produce fraudulent literature or may be used to generate fake scientific research reports.12–18 ChatGPT, launched by OpenAI in November 2022, reached 100 million users by January 2023.16 This growth has never been seen before in any consumer application. The system may readily detect plagiarism and image authenticity or even identify the most appropriate peer reviewers for a particular paper.11–18 These systems may even detect if the text has been written by chatbots. And this is just the beginning. Their unprecedented capability to instantaneously access, summarize, and present, in an intelligible and attractive manner, the available information could be useful for investigators and authors to help in the scientific and editorial process.11–18 This was the understanding of the ICMJE in its new recommendations: these tools are not prohibited but rather their use should be regulated and acknowledged.5 The role of an author to ‘create’ new knowledge and to take responsibility for it (accountability) currently remains limited to human beings (at least for now…). But, please, stay tuned! Supplementary data are not available at European Heart Journal online. All authors declare no conflict of interest for this contribution.
Loading…
Stored only when both papers are in the corpus — 11 citations exist upstream.
Papers that cite the same references — the relational fallback while embeddings are still backfilling.
Imported from OpenAlex (CC0) — nothing here is generated.