Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
Chinese people are already constantly shifting their vocabulary to evade censors. Remember the Winnie-the-pooh fiasco? It will be a test for an AI to stay up-to-date with euphemisms, etc. If the AI learns and adds to it's censorship dictionary, then it's vulnerable to attacks where people deliberately associate normal language with forbidden topics to mess up the social network.
reddit AI Surveillance 1655944973.0 ♥ 11
Coding Result
DimensionValue
Responsibilitygovernment
Reasoningconsequentialist
Policynone
Emotionmixed
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_idd9dij","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"mixed"}, {"id":"rdc_ideuwn6","responsibility":"company","reasoning":"consequentialist","policy":"industry_self","emotion":"resignation"}, {"id":"rdc_idda2ec","responsibility":"government","reasoning":"consequentialist","policy":"none","emotion":"mixed"}, {"id":"rdc_iddc39b","responsibility":"government","reasoning":"deontological","policy":"none","emotion":"outrage"}, {"id":"rdc_iddfwo2","responsibility":"government","reasoning":"consequentialist","policy":"none","emotion":"mixed"} ]