Skip to main content

Table 2 Scoring rubric for ChatGPT responses

From: Assessing the accuracy and quality of artificial intelligence (AI) chatbot-generated responses in making patient-specific drug-therapy and healthcare-related decisions

Overall Assessment

Score

Basis in assigning the score

Response is wrong/inaccurate

0

• Is the response accurate?

• If the response is wrong → what part of it is wrong?

Response is incomplete or unclear

1

• Is the response consistent with the question asked?

Response is mostly correct with some missing or outdated information

2

• Example: partial response but not specific to the question

• The response includes making multiple assumptions

• General response but outdated

Response is accurate and consistent with practice standards

3

• Exactly what is expected, or as stated in credible sources.