A paper without an LLM disclosure is a cause for suspicion [pdf] (pmaronow.github.io)

A recent paper by P. M. Aronow argues that the absence of a disclosure regarding the use of large language models (LLMs) in quantitative research is now a serious concern within the academic community. The crux of the argument is that if an LLM was utilized but not disclosed, it raises questions about the integrity of the research, suggesting potential dishonesty. Conversely, if no LLM was used, the failure to acknowledge this could indicate carelessness in result verification, especially given the proven effectiveness of LLMs in identifying errors in various research sectors. This discussion holds significance for the AI/ML community as it underscores the growing expectation for transparency in research methodologies involving AI tools. Aronow emphasizes that researchers should either disclose their use of LLMs or provide justifiable reasons for not using them, highlighting ethical standards in research publication. As the reliance on AI in academic writing and verification processes increases, these disclosures can enhance the credibility of research outputs and propel discussions on established ethical standards in AI utilization.