Abstract
BACKGROUND: ChatGPT is an open-source artificial intelligence (AI) chatbot that uses deep learning to produce human-like text dialog. Its potential applications in the scientific community are vast; however, its efficacy on performing comprehensive literature searches, data analysis and report writing in aesthetic plastic surgery topics remains unknown. This study aims to evaluate both the accuracy and comprehensiveness of ChatGPT's responses to assess its suitability for use in aesthetic plastic surgery research. METHODS: Six questions were prompted to ChatGPT on post-mastectomy breast reconstruction. First two questions focused on the current evidence and options for breast reconstruction post-mastectomy, and remaining four questions focused specifically on autologous breast reconstruction. Using the Likert framework, the responses provided by ChatGPT were qualitatively assessed for accuracy and information content by two specialist plastic surgeons with extensive experience in the field. RESULTS: ChatGPT provided relevant, accurate information; however, it lacked depth. It could provide no more than a superficial overview in response to more esoteric questions and generated incorrect references. It created non-existent references, cited wrong journal and date, which poses a significant challenge in maintaining academic integrity and caution of its use in academia. CONCLUSION: While ChatGPT demonstrated proficiency in summarizing existing knowledge, it created fictitious references which poses a significant concern of its use in academia and healthcare. Caution should be exercised in interpreting its responses in the aesthetic plastic surgical field and should only be used for such with sufficient oversight. LEVEL OF EVIDENCE IV: This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .