TRV-2026-2988Version 1 · Published

Saved · see the current record

Why this version exists

Certified into the record

The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-2988
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-17T06:02:56.558407Z
status: published
lens: p_space
sector: health
headline: Quality, safety, and readability of large language model responses to patient-derived questions on tooth whitening, resin infiltration, composite bonding, and porcelain veneers
dek: Patients are increasingly consulting large language model (LLM) platforms for information regarding aesthetic dental procedures prior to clinical appointments. This study evaluated the quality, safety, and readability of responses produced by ChatGPT, Gemini, and Meta AI in relation to patient-derived questions. Forty-two patient-derived questions regarding tooth whitening, resin infiltration, composite bonding, and porcelain veneers were presented to each platform. Two calibrated evaluators independently assess…
gain_title: (none)
problem_title: In a study scoring 126 chatbot answers to 42 patient questions on whitening, bonding and veneers, none of ChatGPT, Gemini or Meta AI met the Grade 8 reading target (Flesch-Kincaid 10.2 to 16.5) or all benchmarks for complete, safe answers.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: In a study scoring 126 chatbot answers to 42 patient questions on whitening, bonding and veneers, none of ChatGPT, Gemini or Meta AI met the Grade 8 reading target (Flesch-Kincaid 10.2 to 16.5) or all benchmarks for complete, safe answers.
problem_evidence: None of the platforms met the recommended Grade 8 readability target, with FKGL scores ranging from 10.2 to 16.5. | No platform consistently met all benchmarks for complete, safe, and accessible responses
quick_read: A study in The Saudi Dental Journal put 42 patient-derived questions about tooth whitening, resin infiltration, composite bonding and porcelain veneers to ChatGPT, Gemini and Meta AI. Two evaluators scored all 126 answers for quality, safety and readability. None of the three platforms met the recommended Grade 8 reading level, with Flesch-Kincaid grade levels between 10.2 and 16.5.

ChatGPT scored lowest on quality. Average Global Quality Scores were 3.60 for ChatGPT, 3.98 for Gemini and 4.15 for Meta AI, and ChatGPT trailed both rivals on the rubric total and the Modified DISCERN tool. Gemini and Meta AI differed only on actionability, where Meta AI scored higher.

The authors conclude that no platform consistently delivered complete, safe and accessible answers, and they call for professional oversight when patients use chatbots for guidance before a dental consultation. The study scored the text of answers rather than following patients, so it does not show how people acted on the advice.
limitation: Initial inter-rater agreement between the two evaluators was only slight (pre-consensus kappa 0.011 to 0.191), so final scores depended on consensus discussion to resolve disagreements.
tag: Evidence-backed problem
key_points: Forty-two patient-derived questions covered tooth whitening, resin infiltration, composite bonding and porcelain veneers, each put to ChatGPT, Gemini and Meta AI, giving 126 responses. | Two calibrated evaluators scored every response with a five-domain quality rubric, a five-item Modified DISCERN tool and the Global Quality Score. | Average Global Quality Scores were 3.60 for ChatGPT, 3.98 for Gemini and 4.15 for Meta AI. | ChatGPT scored significantly lower than Gemini and Meta AI on rubric total, Modified DISCERN and Global Quality Score (all adjusted p < 0.005). | Accuracy and patient-centeredness did not differ significantly among the platforms (p = 0.292 and p = 0.071). | No platform reached the recommended Grade 8 reading level; Flesch-Kincaid grade levels ranged from 10.2 to 16.5.
rundown: Readability and comprehension were measured with the Flesch Reading Ease Score and the Flesch-Kincaid Grade Level. Every platform exceeded the Grade 8 target the authors cite as recommended for patient information, so the answers were harder to read than that benchmark.

Statistical comparisons used the Friedman test for repeated measures, followed by paired Wilcoxon signed-rank tests with Bonferroni correction. Rubric total, Modified DISCERN total and Global Quality Score all differed significantly across platforms (p < 0.001 for each).

Meta AI and Gemini outperformed ChatGPT on most quality dimensions relevant to aesthetic dental patient information. Gemini and Meta AI did not differ significantly except on actionability, where Meta AI scored higher.
sources:
- peer_reviewed | The Saudi Dental Journal | https://doi.org/10.1007/s44445-026-00265-6 | 2026-11-16
prev: 0000000000000000000000000000000000000000000000000000000000000000
SHA-256 hash
78d757ff51345dbc61e98410ec4bd875852e6d1ffa984eb515669db02ba3ceed
Previous version’s hash
0000000000000000000000000000000000000000000000000000000000000000
Check this record hasn’t been altered
How to check it yourself, without trusting this page

Download any version from /api/record/TRV-2026-2988 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.