Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
I make applications with AI APIs. One of the features I build-in for customers is watermark removal. Sometimes models use different character sets for the same punctuation in the same copy; so you flatten the charset to just one type. Then, according to the Copyright Office, as long as you change a couple words with synonyms, youre providing enough “human” input to be considered “human” made and copyrightable. You just have to do a couple things to get around it. NBD.
reddit AI Jobs 1683550924.0 ♥ 1
Coding Result
DimensionValue
Responsibilityuser
Reasoningconsequentialist
Policyindustry_self
Emotionapproval
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_jj3pb4f","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"}, {"id":"rdc_jj3scwg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"}, {"id":"rdc_jj413yk","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"}, {"id":"rdc_jj4ckfm","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"indifference"}, {"id":"rdc_jjbxh74","responsibility":"user","reasoning":"consequentialist","policy":"industry_self","emotion":"approval"} ]