Table 3 Categories, examples, and frequencies of educators' concerns regarding ChatGPT and other ADWTs in Research
Answer Category | Examples | F |
|---|---|---|
Ethical Concerns: (e.g., accountability, trust, misuse and abuse, transparency) | Researchers can claim work that is not their own A major concern regarding AI in academic writing is plagiarism. Using ChatGPT without proper citation or attribution of sources can result in academic dishonesty | 26 |
Concerns over the reliability, authenticity, accuracy, and validity of outputs | Many of its suggested sources are not searchable on Google, websites, and databases. It also produces inconsistent data analysis of quantitative data | 25 |
Becoming too dependent on the tools | Researchers might rely on them too often | 11 |
Hinder the development of researchers' skills | The researcher’s skills may not be fully honed as the AI takes the responsibility of completing major writing tasks | 10 |
Becoming impersonal lacking human touch | It lacks the human touch; there are other intricate and idiosyncratic characteristics of human writers which cannot be copied by AI | 7 |
Concerns over the proliferation of AI-written content and turning these tools into publication machines | That articles will be written by AI Since educators are required to publish, many will use such tools just to be able to publish research papers quickly | 7 |
Promoting laziness and complacency | Writers, both students and professors, will become lazy if they rely on these tools | 3 |
Robbing researchers the opportunity to make their meaningful contributions | Allowing AI technology to write what a researcher can and should be able to do themselves robs them of the opportunity to truly make meaningful contributions to their field by their abilities to truly understand the context of the problems they challenge | 1 |
Inequality resulting in disparity in research quality | Unequal access to AI-powered DWTs raises concerns about disparities in research quality and opportunities, as financial resources and technological infrastructure may vary across researchers and institutions | 1 |