Technology - the product of humans’ essence of knowledge - has made everything easier than ever, not only in daily life but also at work. Artificial Intelligence, or AI, affects our information-seeking, decision-making and reduces physical risks in the workplace (University of York). The education market, where AI valued 4 billion USD in 2022 is not an exception (GMI, 2023). Thanks to automation, the workload of teachers can be reduced from 20% to 40%, which is equivalent to a 5-hour cut in their time for lesson preparation (GMI, 2023). AI, therefore, is considered an essential part of future education, according to the majority of educators. In this context, ChatGPT appeared in an undeniably impressive way. With various outstanding features, ChatGPT quickly took the spotlight in recent times, with various outstanding features, especially the ability to process information extremely quickly in any field. Since then, many questions have been raised about whether ChatGPT can replace the role of educators in education when the information it provides is not inferior to any other material but even more optimal. So will “AI” be the educators in the near future? 

 

1. The promise of AI in revolutionizing the education

          The ability of ChatGPT was evaluated on the United States Medical Licensing Exam (USMLE). This expert-level test is the requirement for medical licensure in the United States. ChatGPT is the first to score around the passing mark of 60% accuracy, which has marked a notable milestone in the AI industry (SciTechDaily, 2023). ChatGPT now stands the highest chance of replacing the tech giant Google as the best answer machine. If Google only provides users with relevant websites, ChatGPT gives out responses using the latest AI algorithms. It can not only create content but also write code. Even when users are required to write a song or ask any question in the workbook, this chatbot can solve it quickly. Another factor distinguishing ChatGPT from other chatbots is that it can create a human-like conversation with users. As being trained on a huge data set, ChatGPT is one of the most potent language processing models that can reply in various styles and languages (Euronews.next, 2022).

 

         The integration of AI language models, such as ChatGPT, into educational programs, is increasingly being recognized as a future development in the field of education. Despite the potential benefits and drawbacks, there is no denying that AI language models have the potential to transform the way students learn and teachers teach.

 

       Besides, one of the most significant advantages of AI language models is their ability to provide students with personalized learning experiences. By analyzing a student's strengths, weaknesses, and unique needs, AI models can tailor instruction to meet each requirement. This, in turn, allows for  more customized assistance from teachers, ensuring that learners receive the support they need to succeed. In addition, AI language models can provide quick support to students by providing instant access to answers and resources. This is particularly useful in the research process, where students may require assistance in locating relevant information. The time-saving capabilities of AI language models are also noteworthy, as they can automate administrative tasks such as grading, feedback, and frequently asked questions, freeing up teachers to focus on more challenging tasks (Marr, 2021). Another significant advantage of AI language models is their potential to enhance accessibility for students with disabilities. Interestingly, support tools such as speech recognition, text-to-speech, and Braille translation can help remove barriers to learning, creating a more inclusive educational environment.

 

2. The potential perils of AI in education

        However, as often happens in life, too much of a good thing can be bad. Over-reliance on AI language models can lead to a decrease in critical thinking and problem-solving skills among students. Not to mention, dependence on AI is believed to result in a loss of empathy and personalization, which are considered crucial factors contributing to a supportive and inclusive learning environment. Therefore, failure in preserving such values with this new technology can negatively result in the lack of student engagement and motivation. 

 

         Moreover, AI language models may perpetuate the biases of their developers or the data they are trained on, resulting in discrimination against underrepresented groups, such as people of color, people with disabilities or people from a lower socio-economic status (Humble & Mozelius, 2022). Furthermore, privacy and security concerns must be taken seriously, as AI has the capacity to gather and retain sensitive information about students and educators (Hu et al., 2023). Therefore, while integrating AI language models into educational programs presents potential advantages, it is critical to proceed with caution and carefully assess the associated risks.

 

3. Debates around ChatGPT’s seemingly unpredictable consequences

        Since ChatGPT emerged, there has been lots of concern about the future, where the development of AI diminishes the role of humans. It is believed that we should both harness and hamper the development of these automatic devices. Many people feel a stab of fear that humans will eventually become lazier as the advancement of these technologies reduces the need for students to brainstorm. These concerns are not completely unreasonable. In fact, ChatGPT has various functions, including solving exercises and assignments (Beebom, 2023). Various questions have been raised: Will students really be less active in thinking and more dependent on technology? So is it true that the invention of ChatGPT in particular and technology in general will undermine the abilities of humans?

 

        Regarding these concerns, are we blowing ChatGPT and AI out of proportion? Before ChatGPT, wireless communication technology had brought radical changes to society, especially in education. In the early 20th century, radio emerged as the first electronic mass medium, which delighted the public by delivering information and entertainment in a way never thought possible before (Britannica, n.d.). Here came the efflorescence of knowledge without borders, where humans can catch the news in other places on the spot. Then, 1945 saw the advent of television, which evolved from the model of broadcast radio to bring news and entertainment which is now illustrated by images (Britannica, n.d.). As TV provided more dynamic and lively content, it drew the attention of children and youngsters more to educational channels. The appearance of TV also reduced the education gap as it spread information to every region in one country. Unlike TV which allows us to watch only, the Internet appeared in 1983 and allowed users to not only watch but also proactively discover information on their own (Online Library Learning Center, n.d.). Students can obtain the latest information from various sources. However, that goes along with the fact that they may find answers for every exercise without bothering to work. The public once thought that this answering machine would put an end to students’ self-studying. Regardless of all these innovations through time, educators still find ways to adapt, and the education system still goes forward. So why is there so much concern about ChatGPT ruining human cognitive skills?

​

4. Balancing the scales of promise and peril

         Rather than being too skeptical about the potential threats that AI can bring to education, we should concentrate on finding effective ways to exploit the benefits of AI in education and mitigate its downsides. Firstly, it is crucial to educate students about the limitations of AI and make them aware that it is only a tool to assist their learning and not a replacement for human teachers. Emphasizing critical thinking and problem-solving skills is also important to prevent students from over-relying on AI, and ensure they learn how to think independently and creatively. In addition, it is also of great necessity to encourage collaboration and human interaction to promote empathy and social skills among students, as well as create a more engaging and interactive learning environment.

 

        Furthermore, monitoring and addressing bias and discrimination in AI should be a priority. Close supervision by humans can help ensure that AI language models do not perpetuate any biases that may exist in their training data or programming (Hu et al., 2023). Therefore, before incorporating AI models in the official educational program, teachers and educators should be trained on the effective use of AI in teaching and learning, including how to select appropriate tools, how to interpret AI-generated data, and how to provide personalized support that complements rather than replaces human interaction. 

 

         Through the implementation of these comprehensive and proactive measures, we can effectively harness the vast potential of AI in education, thereby circumventing the potential drawbacks associated with this technology. By doing so, we can envision a future where AI stands as an emblem of the most promising advancements in education, providing a multitude of invaluable benefits to students, educators, and institutions alike.

 

Author: Vu Hanh Trang, Tong Gia Linh

___________________________________________________

REFERENCES 

1. A Brief History of the Internet. (n.d.). https://www.usg.edu/galileo/skills/unit07/internet07_02.phtml#:~:text=January%201%2C%201983%20is%20considered,Protocol%20(TCP%2FIP).

2. ChatGPT and AI: embracing the future of education. (2023, February 24). The University of Sydney. https://www.sydney.edu.au/study/why-choose-sydney/student-life/student-news/2023/02/24/chatgpt-and-ai-embracing-the-future-of-education.html

3. Díaz, D. (2022, March 21). Artificial intelligence and its impact on everyday life - University of York. University of York. https://online.york.ac.uk/artificial-intelligence-and-its-impact-on-everyday-life/#:~:text=From%20morning%20to%20night%2C%20going,%2C%20planning%20and%20information%2Dseeking.

4. Fisher, D. E., Fink, D. G., Noll, A. M., & Fisher, M. J. (1999, July 26). Television (TV) | History, Technology, & Facts. Encyclopedia Britannica. https://www.britannica.com/technology/television-technology

5. Gangwar, A., & Gangwar, A. (2023, January 31). 12 Cool Things You Can Do with ChatGPT. Beebom. https://beebom.com/cool-things-chatgpt/

6. Hu, Y., Kuang, W., Qin, Z., Li, K., Zhang, J., Gao, Y., Li, W., & Li, K. (2023). Artificial Intelligence Security: Threats and Countermeasures. ACM Computing Surveys, 55(1), 1–36. https://doi.org/10.1145/3487890

7. Humble, N., & Mozelius, P. (2022). The threat, hype, and promise of artificial intelligence in education. Discover Artificial Intelligence, 2(1). https://doi.org/10.1007/s44163-022-00039-z

8. Hurst, L. (2022, December 15). ChatGPT: Why the human-like AI chatbot suddenly has everyone talking. Euronews. https://www.euronews.com/next/2022/12/14/chatgpt-why-the-human-like-ai-chatbot-suddenly-got-everyone-talking

9. Marr, B. (2021, July 13). How Is AI Used In Education — Real World Examples Of Today And A Peek Into The Future. Bernard Marr. https://bernardmarr.com/how-is-ai-used-in-education-real-world-examples-of-today-and-a-peek-into-the-future/

10. O’Neill, M. (2023, February 14). The Rise of Artificial Intelligence: ChatGPT’s Stunning Results on the US Medical Licensing Exam. SciTechDaily. https://scitechdaily.com/the-rise-of-artificial-intelligence-chatgpts-stunning-results-on-the-us-medical-licensing-exam/#:~:text=The%20researchers%20tested%20ChatGPT's%20performance,75.0%25%20across%20the%20three%20exams.

11. Roose, K. (2023, January 13). Don’t Ban ChatGPT in Schools. Teach With It. The New York Times. https://www.nytimes.com/2023/01/12/technology/chatgpt-schools-teachers.html

12. School, K. (2023, February 20). How the Internet is Beneficial to Students. 21K School. https://www.21kschool.com/blog/how-the-internet-is-beneficial-to-students/#:~:text=The%20internet%20has%20established%20itself,be%20crucial%20for%20their%20academics 

13. Skretvedt, R., & Sterling, C. H. (2011, September 29). Radio | Definition, History, & Facts. Encyclopedia Britannica. https://www.britannica.com/topic/radio

14. Stanford Accelerator For Learning. (2023, February 15). Provocations and possibilities: Stanford seminar examines the future of ChatGPT in education • Stanford Accelerator for Learning. Stanford Accelerator for Learning. https://acceleratelearning.stanford.edu/story/provocations-and-possibilities-stanford-seminar-examines-the-future-of-chatgpt-in-education/

15. Wadhwani, P., & Gawande, S. (2023, January 23). Artificial Intelligence (AI) in Education Market Size By Component (Solution, Service [Professional Service, Managed Service]), By Deployment (On-premise, Cloud), By Technology (Machine Learning, Deep Learning, Natural Language Processing (NLP)), By Application (Learning Platform & Virtual Facilitators, Intelligent Tutoring System (ITS), Smart Content, Fraud & Risk Management), By End-use (Higher Education, K-12 Education, Corporate Learning), COVID-19 Impact Analysis, Regional Outlook, Growth Potential, Competitive Market Share & Forecast, 2023 – 2032. Global Market Insights Inc. https://www.gminsights.com/industry-analysis/artificial-intelligence-ai-in-education-market