ChatGPT – A risk to academic integrity?

The field of education has been profoundly affected by artificial intelligence (AI), which has changed how students learn and teachers impart knowledge. AI is reshaping conventional teaching and learning as a result of technological advancements, opening up novel opportunities for both teachers and students. Currently, AI is being employed by educational institutions to personalize the learning experience for students.

ChatGPT

Since its launch on November 30th, 2022, the ChatGPT has taken the education industry by storm. It is an AI-powered chatbot developed by Silicon Valley startup OpenAI, which is designed to mimic human-like conversations and generate responses based on user input.  It is trained on a vast dataset of human language, enabling it to provide relevant and informative responses to user queries.

ChatGPT and Higher Education Sector

Several months have passed since its release, and there is compelling evidence to suggest that ChatGPT possesses remarkable capacity to both revolutionize and disrupt the status quo in the education sector. As educational institutions, teachers and students grapple with the opportunities and challenges presented by this bot, a pressing question looms large: does ChatGPT pose a threat to the integrity of education?

When asked this question, here is what ChatGPT has to say.

KEY ISSUES OF INTEGRITY  

ChatGPT can enhance higher education by providing personalized and engaging learning experiences, but its use raises concerns around academic integrity and learning quality.

For example, over-reliance on ChatGPT for attempting assessments may hinder students' development of critical thinking, analytical, and creative writing skills, potentially limiting their ability to cultivate crucial competencies. In February 2023, The American Association of Collegiate Registrars and Admissions Officers (AACRAO) reported a case of students using ChatGPT in exams. Sally Brandon, an Associate Communications Lecturer at Deakin University raised concerns regarding the use of AI to cheat in exams when she discovered 20% of her class’s assessment had used ChatGPT, of which 10 assessments showed evidence of significant bot involvement. Cases as such can potentially have detrimental effects on the student's learning and progress.

Then there is the complex challenge of detecting plagiarism as current plagiarism detection softwares such as Turnitin is still in its infancy stage in effectively identifying content generated by AI. As a result, copying from ChatGPT could potentially evade detection by such software. This places an additional onus on teachers to manually cross-check student responses with those generated by ChatGPT, leading to a heavier workload and potentially reducing the time they have available for other productive work.

Accuracy of information is also a concern with ChatGPT, as it may produce incomplete or erroneous responses due to its training from multiple online platforms. While it can provide detailed responses, this must always be used with caution. Take the case of Dilan Thampapillai, the Associate Dean at the University of New South Wales Business School. In January 2023, The Guardian reported that he had conducted some searches using the chatbot and discovered that it had made mistakes in identifying information across various search results. So, if students fail to verify information they take from ChatGPT, they are likely produce an inaccurate academic piece of work.

The availability of ChatGPT can create academic disparity as some students gain an unfair advantage over their peers who lack access or do not utilize the tool. This impacts both institutions and individuals, as those dependent on ChatGPT may outperform peers, raising ethical concerns about fairness and equity in educational opportunities based on technology use, rather than individual efforts and abilities.

SUPPLEMENT, NOT SUBSTITUTE

Technology and education are intertwined, but reliance on technology like ChatGPT should be as a supplement, not a substitute, to ensure a balanced approach in the progressing landscape of education.

Ethan Mollick, a Professor at the Wharton School at the University of Pennsylvania describes ChatGPT as a “learning companion” with great potential. He has used the bot to assist him to develop a syllabus, lecture, assignment, and grading rubric for MBA students.

Current research and experience of educators around the world illustrate that ChatGPT can be used as a starting-off point for students. For example, some institutions are now using it as an assignment where students generate an essay on ChatGPT and then they edit and fact check it. Some other institutions are encouraging the use of ChatGPT to address writer's block, or to use it to generate examples of what an answer should look like.  

Nonetheless, it is crucial that institutions formulate clear-cut policies and guidelines to govern the use of ChatGPT. There should be strong emphasis on the importance of academic integrity and the need to verify information sourced from the chatbot. As this is an evolving technology, institutions and educators must be vigilant and be trained on the proper to enhance teaching and learning. Take the example of Australia’s The Group of Eight. The group has announced that they have made changes to their assessment methods in response to the emergence of new technologies. Dr. Matthew Brown, the Deputy Chief Executive of the group, stated that their institutions are taking proactive measures to address the use of AI in academic dishonesty by providing education to students, training for staff, revising assessment designs, and implementing targeted detection strategies. This includes revising their assessment approach for 2023, including increased use of supervised exams, a greater emphasis on pen and paper exams and tests, and limiting tests only to units with low integrity risks.

It is only through tackling these concerns head-on that ChatGPT can be utilized efficiently in higher education, without posing a threat to academic integrity.