Evaluation of Artificial Intelligence-Generated Information on Cell Culture and Laboratory Protocols in the Field of Tissue Engineering: A Comparison between GPT-3.5, Claude-Instant, and Microsoft Copilot: AI in Cell Lab
نویسندگان
1 Student Research Committee, School of Dentistry, Shahid Beheshti University of Medical Sciences, Tehran, Iran
2 DDS, Dentofacial Deformities Research Center, Research Institute of Dental Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
3 Endodontic Research Center, Research Institute for Dental Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
4 Department of Tissue Engineering and Applied Cell Sciences, School of Advanced Technologies in Medicine, Shahid Beheshti University of Medical Sciences, Tehran, Iran
5 Assistant Professor, Dental Research Center, Research Institute for Dental Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
doi
10.22037/rrr.v9.45423چکیده
Background and objectives: The current study aimed to evaluate and compare the validity and precision of the information provided by three online artificial intelligence (AI) platforms, including GPT-3.5 (Open AI), Claude-Instant (Anthropic), and Copilot (Microsoft), in the context of laboratory protocols in the field of tissue engineering. Materials and methods: Three specialists in the field of tissue engineering research and molecular and cellular laboratory methods created a survey with 20 open-ended questions. The questions included a variety of subjects regarding cell culture processes and cell analysis techniques. The chatbots' responses to the open-ended inquiries were assessed according to the modified global quality scale. A P-value less than 0.05 was considered statistically significant. Results: GPT-3.5, with a median score of 4 out of 5, performed significantly better than Claude-Instant and Copilot. GPT-3.5 showed superior performance compared to Claude-Instant (P < 0.001) and Copilot (P = 0.007). Both Claude-Instant and Copilot chatbots received a median score of 3 out of 5 in total. Conclusion: In conclusion, AI language models provided satisfactory responses for laboratory procedures in the field of tissue engineering. Specifically, GPT-3.5 outperformed both Claude-Instant and Microsoft Copilot chatbots. Nevertheless, the information offered by all three chatbots was insufficient in terms of the crucial details required for the design and execution of an in vitro analysis.
