Pressure to publish introduces large‐language model risks

Large-language models (LLMs) have the potential to accelerate research in ecology and evolution, cultivating new insights and innovation. However, whilst revelling in the plethora of opportunities, researchers need to consider that LLM use could also introduce risks. An important piece of context underpinning this perspective is the pressure to publish, where research careers are defined, at least partly, by publication metrics like number of papers, impact factor, citations etc. Coupled with academic employment insecurity, especially during early career, researchers may reason that LLMs are a low-risk and high-reward tool for publication. However, this pressure to publish can introduce risks if LLMs are used as a shortcut to game publication metrics instead of a tool to support true innovation. These risks may ultimately reduce research quality, stifle researcher development and incur reputational damage for researchers and the entire scientific record. We conclude with a series of recommendations to mitigate the magnitude of these risks and encourage researchers to apply caution whilst maximising LLM potential. Innovation invites excitement over novel uses, concern over misuses and fears about detrimental impacts on individuals and society. Large-language models (LLMs) represent a significant innovation that could impact how science is conducted, for better and for worse. Cooper et al. (2024) provide a timely overview of LLM use for research and teaching in ecology and evolution and suggest approaches to maximise LLM utility, especially in coding exercises. We agree with the points made by Cooper et al. (2024), but in this complementary extension, we highlight that the potential of LLMs extends beyond coding and could transform the entire research process from writing to reviewing and introduces new risks to scientific progress if applied incautiously. We term these risks: paper hacking, stunted researcher development and reputational risk.

To frame our perspective, an important piece of context is the pressure to publish and the use of publication metrics as markers of researcher accomplishment. Scientists are typically judged through academic publishing and are incentivised to publish to progress in their career, that is ‘publish or perish’ (van Dalen & Henkens, 2012). Indeed, over a 10-year period, researchers beginning their careers in 2000 published 2.6 times more papers than researchers beginning their careers in 1950 (Fire & Guestrin, 2019), with the number of publications rising exponentially across an expanding number of journals (McGill, 2024). Combined with the current global socio-economic climate and academic job rarity, pressure on researchers (especially early career), is high. Against this backdrop of incentivised output and employment insecurity, researchers may reason that LLMs are a valuable tool for increasing publication rates.