Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
I am conflicted. Reports like this make me anxious about my job, but on the other hand, most of the output from AI I've seen in real life largely suck. Not only is it often obvious that a work product is AI-generated (sloppy), it also seems that AI models have a tendency to flat out lie to fill the gaps in its knowledge. It doesn't seem to know when and THAT it can say "I don't know."
youtube AI Jobs 2025-05-30T15:0…
Coding Result
DimensionValue
Responsibilityai_itself
Reasoningmixed
Policyunclear
Emotionfear
Coded at2026-04-26T19:39:26.816318
Raw LLM Response
[ {"id":"ytc_UgxgNUZMwzBtQusQSEZ4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"fear"}, {"id":"ytc_Ugzhrth8mH1ekJivfB94AaABAg","responsibility":"ai_itself","reasoning":"mixed","policy":"unclear","emotion":"fear"}, {"id":"ytc_UgxbTUpAvkcEPK0d0WR4AaABAg","responsibility":"company","reasoning":"deontological","policy":"unclear","emotion":"outrage"}, {"id":"ytc_UgzGPsDnFovgheph6gB4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"liability","emotion":"fear"}, {"id":"ytc_UgzQTwZfGUmZWBxPkZJ4AaABAg","responsibility":"government","reasoning":"unclear","policy":"regulate","emotion":"approval"} ]