Application of artificial intelligence chatbots, including ChatGPT, in education, scholarly work, programming, and content generation and its prospects: a narrative review
2
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
This study aims to explore ChatGPT’s (GPT-3.5 version) functionalities, including reinforcement learning, diverse applications, and limitations. ChatGPT is an artificial intelligence (AI) chatbot powered by OpenAI’s Generative Pre-trained Transformer (GPT) model. The chatbot’s applications span education, programming, content generation, and more, demonstrating its versatility. ChatGPT can improve education by creating assignments and offering personalized feedback, as shown by its notable performance in medical exams and the United States Medical Licensing Exam. However, concerns include plagiarism, reliability, and educational disparities. It aids in various research tasks, from design to writing, and has shown proficiency in summarizing and suggesting titles. Its use in scientific writing and language translation is promising, but professional oversight is needed for accuracy and originality. It assists in programming tasks like writing code, debugging, and guiding installation and updates. It offers diverse applications, from cheering up individuals to generating creative content like essays, news articles, and business plans. Unlike search engines, ChatGPT provides interactive, generative responses and understands context, making it more akin to human conversation, in contrast to conventional search engines’ keyword-based, non-interactive nature. ChatGPT has limitations, such as potential bias, dependence on outdated data, and revenue generation challenges. Nonetheless, ChatGPT is considered to be a transformative AI tool poised to redefine the future of generative technology. In conclusion, advancements in AI, such as ChatGPT, are altering how knowledge is acquired and applied, marking a shift from search engines to creativity engines. This transformation highlights the increasing importance of AI literacy and the ability to effectively utilize AI in various domains of life.
Abstract
This study aims to explore ChatGPT’s (GPT-3.5 version) functionalities, including reinforcement learning, diverse applications, and limitations. ChatGPT is an artificial intelligence (AI) chatbot powered by OpenAI’s Generative Pre-trained Transformer (GPT) model. The chatbot’s applications span education, programming, content generation, and more, demonstrating its versatility. ChatGPT can improve education by creating assignments and offering personalized feedback, as shown by its notable performance in medical exams and the United States Medical Licensing Exam. However, concerns include plagiarism, reliability, and educational disparities. It aids in various research tasks, from design to writing, and has shown proficiency in summarizing and suggesting titles. Its use in scientific writing and language translation is promising, but professional oversight is needed for accuracy and originality. It assists in programming tasks like writing code, debugging, and guiding installation and updates. It offers diverse applications, from cheering up individuals to generating creative content like essays, news articles, and business plans. Unlike search engines, ChatGPT provides interactive, generative responses and understands context, making it more akin to human conversation, in contrast to conventional search engines’ keyword-based, non-interactive nature. ChatGPT has limitations, such as potential bias, dependence on outdated data, and revenue generation challenges. Nonetheless, ChatGPT is considered to be a transformative AI tool poised to redefine the future of generative technology. In conclusion, advancements in AI, such as ChatGPT, are altering how knowledge is acquired and applied, marking a shift from search engines to creativity engines. This transformation highlights the increasing importance of AI literacy and the ability to effectively utilize AI in various domains of life.
참고문헌 (15)
- "ChatGPT 2023 Jan 9 version" (2023)[URL]
- Dezfouli A, "Habits, action sequences and reinforcement learning", Eur J Neurosci, 35, pp.1036~1051 (2012)[DOI]
- "Learning from human preferences" (2017)[URL]
- "Introducing ChatGPT Plus" (2023)[URL]
- 허선, "Are ChatGPT’s knowledge and interpretation ability comparable to those of medical students in Korea for taking a parasitology examination?: a descriptive study", Journal of Educational Evaluation for Health Professions, 20, pp.1~ (2023), 한국보건의료인국가시험원
- Kung TH, "Performance of ChatGPT on USMLE : potential for AI-assisted medical education using large language models", PLOS Digit Health, 2, pp.e0000198~ (2023)[DOI]
- Gao CA, "Comparing scientific abstracts generated by ChatGPT to real abstracts with detectors and blinded human reviewers", NPJ Digit Med, 6, pp.75~ (2023)[DOI]
- Blanco-Gonzalez A, "The role of AI in drug discovery: challenges, opportunities, and strategies", Pharmaceuticals(Basel), 16, pp.891~ (2023)[DOI]
- O’Connor S, "Open artificial intelligence platforms in nursing education: tools for academic progress or abuse?", Nurse Educ Pract, 66, pp.103537~ (2023)[DOI]
- Stokel-Walker C, "ChatGPT listed as author on research papers : many scientists disapprove", Nature, 613, pp.620~621 (2023)[DOI]
- "Can ChatGPT translate?: discover its translation capabilities" (2022)[URL]
- Sun Huh, "Emergence of the metaverse and ChatGPT in journal publishing after the COVID-19 pandemic", Science Editing, 10(1), pp.1~4 (2023), 한국과학학술지편집인협의회[DOI]
- Ouyang L, "Training language models to follow instructions with human feedback", arXiv, ~ (2022)
- "AI and I: the age of artificial creativity" (2022)[URL]
- Carr N., "The shallows: how the Internet is changing the way we think, read and remember" (2020), Atlantic Books