GUEST POST: The Benefits and Risks of ChatGPT for Education
By Maria Radeva
Maria Radeva has been working for the University of Leeds for over two years. She supports undergraduate students in overcoming personal and learning difficulties and enhancing their learning experience in the School of Dentistry. She works closely with academic staff and institutional services to provide students with guidance on academic policy issues and referrals to specialist support services where necessary.
Maria completed her MA in Psychology in 2021 at the University of Dundee, where she undertook a Cognition module led by Dr Carolina Kuepper-Tetzel. Maria is passionate about writing, and her articles, short stories and poems have been published in The Magdalen (student-led), Psych-Talk (led by the British Psychology Society), Marketing Gazette and the Student Publication Association national magazine. You can follow her work on LinkedIn.
This post was originally published on the TILE Network website (here).
(Cover photo by Matheus Bertelli on Pexels)
Rapid technological advancement in the 21st century has influenced many sectors, including education. With over 100 million users, the new Artificial Intelligence tool ChatGPT (Generative Pre-trained Transformer) has gained significant popularity among students, teachers and researchers since its’ release in November 2022 (1). ChatGPT is a chatbot that uses Natural Language Processing to provide human-like answers to users’ queries and engage in natural-sounding conversations (2). ChatGPT is programmed to carry out complex tasks, such as writing articles, poems, essays and even original computer codes, as well as translating, summarizing or expanding texts (3,4). Due to ChatGPT’s efficiency, students report utilizing the tool for academic support and social and personal purposes (5).
The growing presence of ChatGPT in students’ lives has led to debates among educators and changes in educational practices. While some consider ChatGPT and similar AI tools the future of teaching and learning, others perceive them as a threat to developing core skills, such as problem-solving and analytical abilities (6). As a result, several educational institutions have prohibited students from using ChatGPT (7,8). However, incorporating AI in higher education has shown the potential to improve student’s learning capacity and satisfaction (5,9). ChatGPT’s reported advantages and increasing influence on everyday life suggest that prohibiting its use might not be a practical approach. Instead, educators can benefit from examining ChatGPT’s impact on learning and teaching and providing regulations for its use (6). Thus, this article will explore the benefits, drawbacks, and potential implications of utilizing ChatGPT in educational settings.
Image by Andrew Neel from Pexels
Some of the ways ChatGPT can be employed to improve learning and teaching practices include:
Automated Essay Grading: Accumulating evidence suggests that educators can use ChatGPT to automate the grading process of essays, articles, and other forms of written coursework, leaving more time for different aspects of teaching (10). Some studies have found that deep-learning models could grade students’ essays with great accuracy and provide feedback resembling the one produced by human graders (11,12). Therefore, educators can utilise the reports generated by such models to discern the strengths and weaknesses in students’ writing and attend to the areas where students face difficulties. This might allow teachers to target interventions more effectively (10).
Creating Learning Assessment: ChatGPT could assist academic staff and teachers in designing assessments (e.g., quizzes or open-ended prompts) that align with the lesson’s learning objectives and success criteria (6). Since creating quizzes, monthly tests, and examinations is time-consuming, ChatGPT could reduce educators’ workload and improve the quality of questions by using a standard framework (13).
Personalised Tutoring: Evidence suggests that ChatGPT has the potential to serve as a personal tutor to students. Cai et al. (14) used a conversational agent based on a generative model to provide personalised math tutoring to students. The model generated feedback tailored to students’ mistakes, which Cai et al. (14) argued could improve learning outcomes. ChatGPT has the potential to help students with their homework, assignments, projects and even math problems, enabling them to become more autonomous and self-directed learners (15,16). As ChatGPT can generate outlines for articles and other forms of writing, students can use it to help them organise ideas for their research and writing (10).
Interactive Learning: With the assistance of ChatGPT, teachers can design and integrate interactive classroom activities to enhance their pedagogical practices. For example, ChatGPT can create visual aids, such as presentations and worksheets, lesson plans and other educational resources. This can encourage teachers to deliver more dynamic and captivating lessons to meet students’ learning needs (17,18). Rudolph et al. (19) also claimed that educators could utilise ChatGPT’s features to adopt a flipped classroom approach and design creative teaching techniques. This would enable students to study in the classroom and remotely, empowering them to become independent learners.
Image by Shantanu Kumar from Pexels
Despite the numerous advantages of ChatGPT, there are some concerns about its use in educational settings:
Academic integrity issues: Utilising ChatGPT can pose a challenge to academic integrity and credibility. If students fail to cite the use of ChatGPT or the sources in their assignments, this could lead to academic misconduct (20,21). Recent studies found that sophisticated AI tools can bypass traditional plagiarism software like Turnitin. This might encourage students to submit the content produced by ChatGPT as their own (22). In fact, students who have used ChatGPT for their assignments are more likely to plagiarise than those who have not used the tool (23).
Masking learning deficiencies: Some researchers argue that using ChatGPT for educational purposes can hinder educators’ abilities to evaluate student performance accurately (7,24). Educators might struggle to discern the student’s level of understanding of the learning material, resulting in the disguise of learning deficiencies. This can make it more difficult for educators to intervene early on and develop necessary intervention strategies (24).
Ineffective development of core skills: Due to ChatGPT’s efficiency, students can create their work based entirely on its output without using their analytical and decision-making skills (25). Some researchers claim that using ChatGPT can impede the development of essential skills, such as critical thinking, problem-solving, imagination and research abilities (7,10). Therefore, using ChatGPT for educational purposes might negatively affect students’ academic and professional success (25).
Biased data: ChatGPT was trained on a large amount of raw, unpolished data, so some researchers argue against ChatGPT’s reliability and precision (26,27). The quality of ChatGPT’s outputs depends on the diversity of the data it was trained on. If the training datasets include biases, then the content produced by the tool will also be biased. Some biases include overreliance on studies on nations with high income or specific demographics (e.g., white men) and gender biases (28,29).
Inaccurate information: ChatGPT is prone to producing false or misleading information, fabricating articles and using non-existing URLs and references in its responses (6,30). Furthermore, ChatGPT uses only data produced before 2021, making it difficult to rely on its answers for specific topics and recent events (26,31).
Current evidence suggests that using ChatGPT for educational purposes can have a negative impact on academic integrity, student learning and development. However, due to the influence of AI tools on students’ everyday lives, banning ChatGPT might not be a practical solution. Instead, educators should consider integrating the tool into the education system by appropriately adjusting teaching methods and strengthening the examination standards and regulations (6,32). Therefore, assessment practices and institutional protocols should be developed to account for the academic integrity risks imposed by AI tools (25). For example, schools and educational institutions can explore digital-free options in their assignments by designing engaging activities, such as oral presentations, interviews, class debates and group discussions (19,33).
Another solution might be to provide educators with AI-based plagiarism detection tools that can enable them to identify cases of academic integrity (24). Furthermore, educational institutions can train educators to help them understand how to maximise and utilise the potential of ChatGPT in teaching (10). Students should also be taught how to use it effectively to get the maximum benefits of their learning without risking academic malpractice (24). This could be achieved by proactively engaging with students in discussions about ChatGPT. Updating educational practices can equip students with the necessary skills to face society’s future needs (34).
In summary, AI tools like ChatGPT could lessen teachers’ workload and foster interactive and adaptive learning environments. This will leave educators with more resources to attend to students’ learning needs, focus on engaging in professional development, and provide personalized feedback and mentoring to each student (14). However, sophisticated AI tools like ChatGPT also pose challenges to traditional education and teaching, which can limit the development of students’ core skills (25). Therefore, these shortcomings should be addressed to ensure AI tools’ effective and ethical use in educational settings.
References:
(1) Council of the European Union, Analysis and Research Team. ChatGPT in the Public Sector – overhyped or overlooked? (2023). In Council of the European Union. Retrieved October 5, 2023, from https://www.consilium.europa.eu/media/63818/art-paper-chatgpt-in-the-public-sector-overhyped-or-overlooked-24-april-2023_ext.pdf
(2) Deng, J., & Lin, Y. (2023). The Benefits and Challenges of ChatGPT: An Overview. Frontiers in Computing and Intelligent Systems, 2(2), 81–83. https://doi.org/10.54097/fcis.v2i2.4465
(3) Tate, T., Doroudi, S., Ritchie, D., Xu, Y., & Warschauer, M. (2023). Educational Research and AI-Generated Writing: Confronting the Coming Tsunami. EdarXiv Preprints. https://doi.org/10.35542/osf.io/4mec3
(4) Williams, C. (2023). Hype, or the future of learning and teaching? 3 Limits to AI’s ability to write student essays – Kent Academic Repository. The London School of Economics and Political Sciences internet blog. Retrieved October 9, 2023, from https://kar.kent.ac.uk/id/eprint/99505
(5) Forman, N., Udvaros, J., & Avornicului, M. S. (2023). ChatGPT: A new study tool shaping the future for high school students. International Journal of Advanced Natural Sciences and Engineering Researches, 7(4), 95–102. https://doi.org/10.59287/ijanser.562
(6) Baidoo-Anu, D., & Ansah, L. O. (2023). Education in the Era of Generative Artificial Intelligence (AI): Understanding the potential benefits of ChatGPT in promoting teaching and learning. Social Science Research Network. https://doi.org/10.2139/ssrn.4337484
(7) Sullivan, M., Kelly, A., & McLaughlan, P. (2023). ChatGPT in higher education: Considerations for academic integrity and student learning. Journal of Applied Learning and Teaching, 6(1). https://doi.org/10.37074/jalt.2023.6.1.17
(8) Elsen-Rooney, M. (2023, January 3). NYC education department blocks ChatGPT on school devices, networks. Chalkbeat New York. Retrieved October 11, 2023, from https://ny.chalkbeat.org/2023/1/3/23537987/nyc-schools-ban-chatgpt-writing-artificial-intelligence
(9) Opara, E., Theresa, A., & Aduke, T. C. (2023). ChatGPT for Teaching, Learning and Research: Prospects and Challenges. Glob Acad J Humanit Soc Sci, 5(2), 33–40. https://ssrn.com/abstract=4375470
(10) Kasneci, E., Seßler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser, U., Groh, G., Günnemann, S., Hüllermeier, E., Krusche, S., Kutyniok, G., Michaeli, T., & Nerdel, C. (2023). ChatGPT for Good? On Opportunities and Challenges of Large Language Models for Education. EdArXiv Preprints. https://doi.org/10.35542/osf.io/5er8f
(11) Liang, G., On, B., Jeong, D., Kim, H., & Choi, G. S. (2018). Automated essay scoring: A Siamese bidirectional LSTM neural network architecture. Symmetry, 10(12), 682. https://doi.org/10.3390/sym10120682
(12) Lu, C., & Cutumisu, M. (2021). Integrating Deep Learning into An Automated Feedback Generation System for Automated Essay Scoring. International Educational Data Mining Society. Retrieved October 15, 2023, from https://files.eric.ed.gov/fulltext/ED615567.pdf
(13) Zhai, X. (2023). ChatGPT for next generation Science Learning. Social Science Research Network. https://doi.org/10.2139/ssrn.4331313
(14) Cai, W., Grossman, J., Lin, Z., Sheng, H., Wei, J. T., Williams, J. J., & Goel, S. (2021). Bandit algorithms to personalize educational chatbots. Machine Learning, 110(9), 2389–2418. https://doi.org/10.1007/s10994-021-05983-y
(15) Mhlanga, D. (2023). Open AI in Education, The responsible and ethical use of ChaTGPT towards lifelong Learning. Social Science Research Network. https://doi.org/10.2139/ssrn.4354422
(16) Qadir, J. (2022). Engineering Education in the Era of ChatGPT: Promise and Pitfalls of Generative AI for Education. TechRxiv. https://doi.org/10.36227/techrxiv.21789434.v1
(17) Atlas, S. (2023). ChatGPT for Higher Education and Professional Development: A Guide to Conversational AI. Digital Commons, University of Rhode Islans. https://digitalcommons.uri.edu/cgi/viewcontent.cgi?article=1547&context=cba_facpubs
(18) Herft Educator. (2023). A Teacher’s Prompt Guide to ChatGPT aligned with “What Works Best” Guide. User Generated Education. https://usergeneratededucation.files.wordpress.com/2023/01/a-teachers-prompt-guide-to-chatgpt-aligned-with-what-works-best.pdf
(19) Rudolph, T., Tan, S., & Tan, S. (2023). ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? Journal of Applied Learning and Teaching, 6(1). https://doi.org/10.37074/jalt.2023.6.1.9
(20) Kleebayoon, A., & Wiwanitkit, V. (2023). Artificial intelligence, chatbots, plagiarism and basic honesty: comment. Cellular and Molecular Bioengineering, 16(2), 173–174. https://doi.org/10.1007/s12195-023-00759-x
(21) Thurzo, A., Strunga, M., Urban, R., Surovková, J., & Afrashtehfar, K. I. (2023). Impact of Artificial Intelligence on Dental Education: A Review and Guide for curriculum update. Education Sciences, 13(2), 150. https://doi.org/10.3390/educsci13020150
(22) Khalil, M., & Er, E. (2023). Will ChatGPT get you caught? Rethinking of Plagiarism Detection. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2302.04335
(23) Basic, Z., Banovac, A., Kruzic, I., & Jerkovic, I. (2023). Better by you, better than me, ChatGPT-3.5 as writing assistance in students’ essays. arXiv, 2. https://doi.org/10.48550/arXiv.2302.04536
(24) Grassini, S. (2023). Shaping the Future of Education: Exploring the potential and consequences of AI and ChatGPT in educational settings. Education Sciences, 13(7), 692. https://doi.org/10.3390/educsci13070692
(25) Sok, S., & Heng, K. (2023). ChaTGPT for Education and Research: A Review of Benefits and Risks. Social Science Research Network. https://doi.org/10.2139/ssrn.4378735
(26) Rahimi, F., & Abadi, A. T. B. (2023). ChatGPT and publication ethics. Archives of Medical Research, 54(3), 272–274. https://doi.org/10.1016/j.arcmed.2023.03.004
(27) Sallam, M. (2023). ChatGPT Utility in Healthcare Education, Research, and Practice: Systematic Review on the promising perspectives and valid concerns. Healthcare, 11(6), 887. https://doi.org/10.3390/healthcare11060887
(28) Mbakwe, A. B., Lourentzou, I., Celi, L. A., Mechanic, O. J., & Dagan, A. (2023). ChatGPT passing USMLE shines a spotlight on the flaws of medical education. PLOS Digital Health, 2(2), e0000205. https://doi.org/10.1371/journal.pdig.0000205
(29) Lucy, L., & Bamman, D. (2021). Gender and Representation Bias in GPT-3 Generated Stories. Proceedings of the Third Workshop on Narrative Understanding, 48–55. https://doi.org/10.18653/v1/2021.nuse-1.5
(30) Gordijn, B., & Have, H. T. (2023). ChatGPT: evolution or revolution? Medicine Health Care and Philosophy, 26(1), 1–2. https://doi.org/10.1007/s11019-023-10136-0
(31) Hassani, H., & Silva, E. S. (2023). The role of ChatGPT in Data Science: How AI-Assisted conversational Interfaces are revolutionizing the field. Big Data and Cognitive Computing, 7(2), 62. https://doi.org/10.3390/bdcc7020062
(32) Lo, C. K. (2023). What is the impact of CHATGPT on education? A rapid review of the literature. Education Sciences, 13(4), 410. https://doi.org/10.3390/educsci13040410
(33) Cotton, D., Cotton, P. A., & Shipway, J. R. (2023). Chatting and cheating: Ensuring academic integrity in the era of ChatGPT. Innovations in Education and Teaching International, 1–12. https://doi.org/10.1080/14703297.2023.2190148
(34) Huang, Y. (2023). Reflection on whether Chat GPT should be banned by academia from the perspective of education and teaching. Frontiers in Psychology, 14. https://doi.org/10.3389/fpsyg.2023.1181712
