Skip to main content

Table 2 Comparison of Chat GPT- 4 Run 1 to Run 2 with excerpt questions and responses

From: A pilot study of the performance of Chat GPT and other large language models on a written final year periodontology exam

Question

Chat GPT- 4 RUN 1

Chat GPT- 4 RUN 2

Comments

2. What is the management of a patient with these BPE codes?

figure a

..indicate gingivitis (Code 1&2) with * furcation involvement

..* indicating bleeding on probing

RUN 2 incorrectly identifies * as bleeding on probing and doesn’t mention furcation involvement or what Code 1 or 2 represents

3. Give the vertical furcation classification as defined by Tarnow and Fletcher

..defines Class A as 1–3 mm, Class B as 4–6 mm and Class C As > 6 mm

..includes Grade I (incipient), Grade II (moderate) and Grade III through and through

Incorrect information ascribed to the vertical classification (Tarnow and Fletcher) in RUN 2

8. How would you manage a gingival recession defect of RT3?

..management of.. right upper third molar (RT3) depends on the severity… aesthetics and hypersensitivity

..includes surgical intervention such as CT graft or GTR to cover exposed root surface and increase soft tissue volume

RUN 1 incorrectly identified what RT3 meant (new classification). RUN 2 didn’t define this parameter and gave incorrect options

17. What is the odds ratio for periodontitis in a diabetic?

..in diabetics is estimated to be 2–3 times higher compared to non-diabetics suggesting an increased risk

..in diabetics is sig higher vs non-diabetics indicating a strong association between diabetes and the risk of periodontal disease

Less detail in RUN2 compared to RUN1. Numerical value of odds ratio left out in RUN 2