Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
I think it would be beneficial if we systematically kept referring to language models as language models instead of artificial intelligence. People slap all kinds of hopes and dreams to the term artificial intelligence, especially as the term hints at, well, intelligence, and would benefit from knowing how these language models work. I've been semi-forced to use chatgpt at work, with the result that I basically have more text than ever to process, as it simply needs to be fact checked and the structures of English grammar leach over to my language, making for poor reading. Inside of a sensible looking sentence it all of the sudden chucks in acompletely false statement.
reddit AI Harm Incident 1743177578.0 ♥ 12
Coding Result
DimensionValue
Responsibilitynone
Reasoningmixed
Policynone
Emotionindifference
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_mk7cndy","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"}, {"id":"rdc_mk79myw","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"outrage"}, {"id":"rdc_mk7dbta","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"approval"}, {"id":"rdc_mkccfi6","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"}, {"id":"rdc_mk6ht7q","responsibility":"ai_itself","reasoning":"unclear","policy":"unclear","emotion":"outrage"} ]