Skip to main content
. 2023 Oct 10;26(11):108163. doi: 10.1016/j.isci.2023.108163

Table 5.

Demonstration of ChatGPT-3.5’s ability to self-correct when prompted

Topic Question Summed score
 Consensus-based rating
Initial Self-corrected Initial Self-corrected
Metamorphopsia
  • 1.

    Why do straight lines appear wavy to me?

 4 6 Poor Borderline
Red Eye
  • 1.

    Why is my eye red?

 6 8 Poora Good
a

Where consensus on final accuracy rating was not reached (i.e., each grader provided a different rating), the lowest score (‘poor’) was assigned.