The advantages and limitations of using ChatGPT to enhance technological research

被引:46
作者
Rice, Stephen [1 ]
Crouse, Sean R. [1 ]
Winter, Scott R. [1 ]
Rice, Connor [2 ]
机构
[1] Embry Riddle Aeronaut Univ, 1 Aerosp Blvd, Daytona Beach, FL 32114 USA
[2] Univ Florida, Gainesville, FL 32611 USA
关键词
ChatGPT; Artificial intelligence; Research;
D O I
10.1016/j.techsoc.2023.102426
中图分类号
D58 [社会生活与社会问题]; C913 [社会生活与社会问题];
学科分类号
摘要
In 2022, OpenAI made a groundbreaking entrance with the release of ChatGPT, a new online chatbot that allows users to interact with the GPT-3.5 language model. Users can ask questions and converse with ChatGPT by typing into a text field similar to direct messaging software. Then, ChatGPT will generate a response. Users can then either respond to ChatGPT, regenerate the previous response, or "like" the response and give feedback. OpenAI improved the program on March 14th, 2023, with the release of GPT 4, which promised better reasoning ability. Both these iterations of ChatGPT have attracted significant attention from researchers due to the software's remarkably enhanced capabilities compared to earlier versions. Due to its inherent value as a research tool, ChatGPT will likely become a permanent fixture, so a thorough evaluation of ethical and professional boundaries is crucial. In this opinion paper, we explore ChatGPT 4.0 by addressing: a) its capabilities, b) its limitations and weaknesses, and c) strategies for fact-checking its output to ensure high-quality responses. Subsequently, the authors delve into the diverse implications of this software and discuss how it can be optimally employed to advance research in technology and various other domains.
引用
收藏
页数:7
相关论文
共 24 条
  • [1] Aydin O., 2023, IS CHATGPT LEADING G, DOI [10.2139/ssrn.4341500, DOI 10.2139/SSRN.4341500]
  • [2] Large language models and the perils of their hallucinations
    Azamfirei, Razvan
    Kudchadkar, Sapna R.
    Fackler, James
    [J]. CRITICAL CARE, 2023, 27 (01)
  • [3] Bagde H., 2023, A Systematic Review and Meta-Analysis on Chatgpt and its Utilization in Medical and Dental Research, DOI [10.2139/ssrn.4468440, DOI 10.2139/SSRN.4468440]
  • [4] Belouadi J, 2023, Arxiv, DOI arXiv:2212.10474
  • [5] On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?
    Bender, Emily M.
    Gebru, Timnit
    McMillan-Major, Angelina
    Shmitchell, Shmargaret
    [J]. PROCEEDINGS OF THE 2021 ACM CONFERENCE ON FAIRNESS, ACCOUNTABILITY, AND TRANSPARENCY, FACCT 2021, 2021, : 610 - 623
  • [6] Use of ChatGPT in academia: Academic integrity hangs in the balance
    Bin-Nashwan, Saeed Awadh
    Sadallah, Mouad
    Bouteraa, Mohamed
    [J]. TECHNOLOGY IN SOCIETY, 2023, 75
  • [7] Borji A, 2023, Arxiv, DOI [arXiv:2302.03494, 10.48550/arxiv.2302.03494, DOI 10.48550/ARXIV.2302.03494]
  • [8] Chen YR, 2023, Arxiv, DOI [arXiv:2212.10522, 10.48550/ARXIV.2212.10522, DOI 10.48550/ARXIV.2212.10522]
  • [9] Chatting and cheating: Ensuring academic integrity in the era of ChatGPT
    Cotton, Debby R. E.
    Cotton, Peter A. A.
    Shipway, J. Reuben
    [J]. INNOVATIONS IN EDUCATION AND TEACHING INTERNATIONAL, 2024, 61 (02) : 228 - 239
  • [10] ChatGPT for (Finance) research: The Bananarama Conjecture
    Dowling, Michael
    Lucey, Brian
    [J]. FINANCE RESEARCH LETTERS, 2023, 53