The emergence of new systems based on Large Language Models (LLMs) such as OpenAI’s ChatGPT has created a stir in academia. Since LLMs generate mostly relevant (yet sometimes wrong) outputs in response to provided prompts, they can support researchers in writing tasks. Given the significance of scholarly writing in how research ideas/methods/results are communicated and understood, a discussion about LLMs in scholarly publishing by/with metascience experts is urgent, especially given the current misinformation crisis.

After the generation of first scholarly outputs with LLMs, academic journals/institutions are preparing to define/specify the norms of responsible use of LLMs. This panel discussion will examine challenges and opportunities of using LLMs in research, particularly their employment in writing manuscripts and peer-reviews. The panel will also explore the impact of LLMs on existing (and anticipated) ethical issues in scholarship. The panel will be led by Mohammad Hosseini (Associate Editor, Accountability in Research).