Ruiz A M, Kraus M B, Arendt K W, Schroeder D R, Sharpe E E
Department of Anesthesiology and Perioperative Medicine, Mayo Clinic, Rochester, MN, United States.
Department of Anesthesiology and Perioperative Medicine, Mayo Clinic, Phoenix, AZ, United States.
Int J Obstet Anesth. 2025 Feb;61:104293. doi: 10.1016/j.ijoa.2024.104293. Epub 2024 Nov 15.
A personal statement is a common requirement in medical residency and fellowship applications. Generative artificial intelligence may be used to create a personal statement for these applications.
Two personal statements were created using OpenAI's Chat Generative Pre-trained Transformer (ChatGPT) and two applicant-written statements were collected. A survey was sent to obstetric anesthesia fellowship program directors in the United States to assess the perceived readability, authenticity, and originality of the four personal statements. In addition, the survey assessed perceptions of applicants who use artificial intelligence to write a personal statement, including their integrity, work ethic, reliability, intelligence, and English proficiency.
Surveyed fellowship directors could not accurately discern whether statements were applicant-written or artificial intelligence-generated. The artificial intelligence-generated personal statements were rated as more readable and original than the applicant-written statements. Most program directors were moderately or extremely concerned about the applicant's integrity, work ethic, and reliability if they suspected the applicant utilized ChatGPT.
Program directors could not accurately discern if the statements were written by a person or artificial intelligence and would have concerns about an applicant suspected of using artificial intelligence. Medical training programs may benefit from outlining their expectations regarding applicants' use of artificial intelligence.
个人陈述是医学住院医师和专科培训申请中的常见要求。生成式人工智能可用于为这些申请创建个人陈述。
使用OpenAI的聊天生成预训练变换器(ChatGPT)创建了两份个人陈述,并收集了两份申请人撰写的陈述。向美国产科麻醉专科培训项目主任发送了一项调查,以评估这四份个人陈述的可读性、真实性和原创性。此外,该调查评估了对使用人工智能撰写个人陈述的申请人的看法,包括他们的诚信、职业道德、可靠性、智力和英语水平。
接受调查的专科培训主任无法准确辨别陈述是由申请人撰写还是由人工智能生成。人工智能生成的个人陈述比申请人撰写的陈述被评为更具可读性和原创性。如果大多数项目主任怀疑申请人使用了ChatGPT,他们会对申请人的诚信、职业道德和可靠性感到中度或极度担忧。
项目主任无法准确辨别陈述是由人还是人工智能撰写,并且会对涉嫌使用人工智能的申请人感到担忧。医学培训项目可能会受益于概述他们对申请人使用人工智能的期望。