Skip to main content
. 2023 Dec 25;25:e51229. doi: 10.2196/51229

Table 3.

Accuracy of the judgement of abstracts generated by humans or ChatGPT.

Abstract Accuracy, % Proportion, n/N
Structured abstract with 4 subheadings 83.75 67/80
Structured abstract with 3 subheadings 96.28 77/80
Unstructured abstract 100 80/80