The implication would be that GPT-4.5 was not judged to be human 27% of the time... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		dwpdwpdwpdwpdwp 14 days ago \| parent \| context \| favorite \| on: AI users whose lives were wrecked by delusion The implication would be that GPT-4.5 was not judged to be human 27% of the time. You can't determine how often humans were judged correctly as humans from that data point.

jmalicki 14 days ago [–]

The structure of the test was that there was one human and one AI conversation partner, and the rater had to choose which one was which.

Given that structure, you can judge from that data point.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact