alt Hacker NewsI had chatgpt censor it's initial answer the other day when asking about two cognate words with no sensitivity issues. I asked why I couldn't ask such questions and it relented. How will they know what is provocative?
I had chatgpt censor it's initial answer the other day when asking about two cognate words with no sensitivity issues. I asked why I couldn't ask such questions and it relented. How will they know what is provocative?
The censor is separate from the generation. It didn't "relent" so much as "produce a slightly different output that didn't trigger the censor's threshold."