Neither Artificial, nor Intelligent: Critical Responses to Text Generation by Means of the Artificial Intelligence App ChatGPT

This study aims to analyse critical responses to the generation of texts, including scientific and medical articles by means of Artificial Intelligence. It is argued that the affordances of text-generation models using Artificial Intelligence (AI) are increasingly plausible, as the programs reproduce the format of texts memorized for the purposes of producing computer-generated content, mimicking but not replicating human intelligence. The most recent incarnation of such programs, ChatGPT, a Large Language Model (LLM), has been widely adopted in the academic community, not just by novice writers, since its launch in November 2022, giving rise to considerable alarm as it is evident that the model can replicate essay writing and research articles in a plausible manner, resulting in significant concerns among the editors of scientific journals, including medical journals, dismayed by the large-scale submission of articles drafted not by the purported authors, but by means of ChatGPT. Journal editors are seeking to adopt defence mechanisms to identify and weed out the manuscripts of dubious authorship, but preliminary studies show that they are not readily detectable even by experienced journal editors without adopting AI output detector software. In a critical discourse analysis framework (Bhatia 1993, 2014, 2017) this study examines critical comments from Nature Briefing, Science, the New York Times and the Financial Times, in addition to the scientific publications listed in the Bibliography, in which scholars in a range of disciplines express their concern about the impact of AI-generated text not just on undergraduate essay writing, but also on scientific and medical writing, resulting in plausible sounding text that presents fabricated results that are purported to be authentic, a phenomenon known as hallucination. The critical responses are classified according to author stance, with the authors identified as ALARMISTS, ENTHUSIASTS, SCEPTICS, PROHIBITIONISTS, PHILOSOPHERS, ETHICISTS AND PRAGMATISTS. The article concludes by arguing that ChatGPT should not be viewed as a matter of concern solely for IT specialists as this disruptive technology is raising questions that need to be addressed by language specialists, discourse analysts and scholars concerned about academic integrity and the quality and reliability of scientific and medical discourse.

BIBLIOGRAPHY
Agomuoh F. ChatGPT: how to use the viral AI chatbot that took the world by storm. Digital Trends. Published 13 December 2022. Accessed 6 January 2023. https://www.digitaltrends.com/computing/how-to-use-openai-chatgpt-text-generation-chatbot/
Bhatia, Vijay K. 1993. Analysing Genre: Language Use in Professional Settings. London: Longman.
Bhatia, Vijay K. 2014. Worlds of Written Discourse: A genre-based view. London: Continuum.
Bhatia, Vijay K. 2017. Critical Genre Analysis: Investigating Interdiscursive Performance in Professional Contexts. London: Routledge
Bishop JM. Artificial intelligence is stupid and causal reasoning will not fix it. Front Psychol. 2020;11:513474. Accessed 6 January 2023.
https://pubmed.ncbi.nlm.nih.gov/33584394/#article-details
Briganti G, Le Moine O. Artificial Intelligence in Medicine: Today and Tomorrow. Front Med (Lausanne). 2020;7:27. Accessed 8 February 2023
https://www.frontiersin.org/articles/10.3389/fmed.2020.00027/full.
Clark E, August T, Serrano S, Haduong N, Gururangan S, Smith NA. All that’s ‘human’ is not gold: Evaluating human evaluation of generated text. In: Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Vol. 1: Long Papers). Association for Computational Linguistics; 2021:7282–7296. Accessed