Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
No, 99.5% is a complete failure for systems that will check and recheck everyone. Even 99.9% is a failure, you can’t have people being harassed for no reason. Plus, I’m sure there is a segment of people for which the chine learning algorithm doesn’t work as well for, and given the weird and opaque way neural networks reason youll quickly get some Kafkaesque results of someone who gets binged every day due to the spacing of his eyes, or width of his nostrils together with an earring.
reddit AI Harm Incident 1562202870.0 ♥ 9
Coding Result
DimensionValue
Responsibilitynone
Reasoningconsequentialist
Policynone
Emotionoutrage
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_esqhzpk","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"outrage"}, {"id":"rdc_esqiuqr","responsibility":"none","reasoning":"consequentialist","policy":"ban","emotion":"outrage"}, {"id":"rdc_esqeiei","responsibility":"none","reasoning":"consequentialist","policy":"industry_self","emotion":"indifference"}, {"id":"rdc_espuyj8","responsibility":"none","reasoning":"consequentialist","policy":"ban","emotion":"fear"}, {"id":"rdc_esq6z4o","responsibility":"none","reasoning":"consequentialist","policy":"regulate","emotion":"fear"} ]