Share this article
Recently, there have been several incidents in which papers published in academic peer-reviewed journals are evident to use Artificial Intelligence and Chatbots in generating parts of the text submitted for publication in peer-reviewed academic journals.
Even worse, some author teams had literally copied and pasted the response of ChatGPT or any other Chatbot into the text and ironically this error was never unmasked neither by the authors, the reviewers nor the editor(s), even if we set aside the role of the professional copy editor usually hired by the publisher to ensure the quality of accepted manuscripts and that the articles are published only when they are free of typos and other common mistakes.
Let us explore the future trends in AI-driven academic research in detail.
Future trends in Artificial Intelligence (AI)
I am not sure how this could happen and I am in deep surprise that so many people failed to detect direct phrases that show clearly that this phrase was copied from a conversation with a Chatbot. What doubled my surprise that such phrases were sometimes the very first sentence in the entire paper and can be read clearly once you decide to spend a couple of seconds reading the few very first words of the abstract.
Chat GPT: Revolutionizing conversational AI and its impact on communication
The debate to me is not about whether AI language models are usable in research or not. Yet, the main issue is to highlight key areas where the interaction of the human element with the research inquiry is essential and hence using AI to bypass some time-consuming yet valuable steps in the research process only reduces the value of the research and the researcher alike.
Sometimes using extreme examples is helpful in identifying the main problem. So if a Ph.D. student is starting to work for his degree 15 years from now and AI language models by then are expected to reach a much more powerful status compared to the present time.
Therefore, if a PhD student basically asks AI to find a research question, summarize the literature, recommend observable research gaps, suggest a suitable research design, recommend how the data should be analyzed and suggest what are the main conclusions, takeaways, theoretical and practical implications of such findings. In that case are we sure that this student has met the objectives of the program of study and should be awarded the degree of Doctor of Philosophy just for acting as an assembler of the input provided by a machine towards a specific topic?
One might argue it’s an extreme case and that maybe only experienced researchers could make use of AI and language models in research because they have enough experience to evaluate the input provided by a Chatbot like ChatGPT and whether it is scientifically valid or not. However, less experienced researchers such as Ph.D. students should not be taking the privilege of using AI because they are not equipped and they have to learn a lot of things by themselves to build a solid research experience.
This argument is problematic for two reasons. First, I am not sure if it is possible for an experienced researcher who accepts the usage of AI in research to convince a Ph.D. student why this should not be used since the senior faculty would have decided to use AI to save more time and to be more productive and thus, the experienced researcher or the supervisor has already, indirectly, emphasized to his students that productivity and the quantity of research output is more important than academic honesty and the quality of the research output.
Second, there will never be a consensus on what defines an experienced researcher from a non-experienced researcher due to the huge variations in the methods used to assess research quality and thus research experience.
Conclusion
In conclusion, it is very important to emphasize that in research the most valuable elements are the research questions we ask the human interpretation of the extant literature, and the results we get out of our analysis. Those appear as the most valuable contributions to knowledge any research effort provides. It seems that we have to respect and defend those two elements so that they remain as human as possible, at least for the next ten years.
