Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-0265
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-07-19T00:57:54.025735Z
status: published
lens: trace
sector: health
headline: Comparison of artificial intelligence-based chatbots and expert periodontists in responding to patient questions: a multi-dimensional analysis
dek: Large Language Model (LLM) -based chatbots are increasingly used in patient information processes. The aim of this study was to compare the performance of ChatGPT (GPT-5.1), Gemini (2.5 Flash), and Claude (Sonnet 4.5) with expert periodontologists in responding to periodontal questions. Responses were evaluated in terms of scientific accuracy, completeness, conciseness & focus, empathy, and clarity, and differences among groups were investigated. The question pool was developed de novo based on clinical experien…
gain_title: LLM chatbots answered periodontal patient questions with scientific accuracy comparable to expert periodontologists while scoring higher on completeness and empathy.
problem_title: Some chatbot models showed lower conciseness & focus and clarity, producing longer less-focused answers that may make it harder for patients to maintain focus and perceive information in a structured manner.
trace_subject: LLM chatbots answering periodontal patient questions for patient education
gain_reading: LLM chatbots answered periodontal patient questions with scientific accuracy comparable to expert periodontologists while scoring higher on completeness and empathy.
gain_evidence: LLM-based chatbot responses achieving higher scores than expert responses | all LLM-based chatbot responses scoring higher than expert responses
problem_reading: Some chatbot models showed lower conciseness & focus and clarity, producing longer less-focused answers that may make it harder for patients to maintain focus and perceive information in a structured manner.
problem_evidence: This may make it more difficult for patients to maintain focus and to perceive information in a structured manner when using AI-generated responses.
quick_read: A July 2026 peer-reviewed study compared ChatGPT GPT-5.1, Gemini 2.5 Flash, and Claude Sonnet 4.5 against expert periodontologists on 20 periodontal patient questions. Nine blinded periodontologists rated anonymized answers for scientific accuracy, completeness, conciseness & focus, empathy, and clarity.

The results matter because patients increasingly use chatbots for dental information. Comparable accuracy with higher completeness and empathy suggests a supportive role for AI in patient education, but lower conciseness in some models and the stated need for physician supervision highlight risks of information overload and unstructured delivery that could impair patient comprehension.
limitation: Findings based on 20 de novo questions and first responses to a standardized prompt, with model differences affecting conciseness and clarity, requiring physician oversight before clinical use.
tag: Automated dual reading
key_points: Compared ChatGPT (GPT-5.1), Gemini (2.5 Flash), and Claude (Sonnet 4.5) to three experienced periodontologists on 20 open-ended periodontal questions. | Nine independent periodontologists blindly rated anonymized responses on 5-point Likert scales for accuracy, completeness, conciseness & focus, empathy, and clarity. | No significant difference in scientific accuracy (p = 0.425); chatbots scored higher on completeness (p < 0.0001) and empathy (p < 0.0001). | Gemini scored lower on conciseness & focus, and clarity differed only between ChatGPT and Gemini, with experts using fewer and more focused expressions.
rundown: Researchers created 20 open-ended periodontal questions validated by Lawshe's method, collected expert answers from three periodontologists, and submitted the same standardized prompt to three LLMs, recording first responses.

All responses were anonymized and scored by nine independent periodontologists using Friedman test with Bonferroni correction; inter-rater agreement was within a good range across all domains.
sources:
- peer_reviewed | BMC Oral Health | https://doi.org/10.1186/s12903-026-09315-1 | 2026-07-17
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

246b6e053f2fec6b65ee6cdbb86b19b5988ae30ac9e8b2295da753d5f3e47392

previous

0000000000000000000000000000000000000000000000000000000000000000