Use of large language models in clinical periodontology

Researchers have examined the performance of large language models when accessing clinical periodontology content.

In a study published in The Journal of Prosthetic Dentistry, the researchers posed 10 common, open-ended clinical questions on periodontology to four large language models: ChatGPT 4.0, Google Gemini, Google Gemini Advanced and Microsoft Copilot. They then asked two periodontists to evaluate the comprehensiveness, scientific accuracy, clarity and relevance of the models’ responses.

The researchers discovered that ChatGPT 4.0 scored the highest, whereas Google Gemini scored the lowest. ChatGPT 4.0’s responses were also found to be comprehensive, scientifically accurate, coherent and relevant.

The findings uncovered the potential risks of relying on inaccurate responses from large language models or improper use of the artificial intelligence tools, highlighting the need for dental professionals to use their clinical expertise when making decisions. The researchers concluded that although the models performed well, AI should only be complementary tools.

Read more: The Journal of Prosthetic Dentistry

The article presented here is intended to inform you about the broader media perspective on dentistry, regardless of its alignment with the ADA's stance. It is important to note that publication of an article does not imply the ADA's endorsement, agreement, or promotion of its content.