Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
If you understand how to instruct it properly it can generate novel outcomes. But I suppose that covers the 10% you are talking about. Most people have no idea how to interact with LLMs properly.
reddit AI Governance 1748740797.0 ♥ 3
Coding Result
DimensionValue
Responsibilityuser
Reasoningunclear
Policyunclear
Emotionindifference
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_mvc486j","responsibility":"government","reasoning":"deontological","policy":"regulate","emotion":"outrage"}, {"id":"rdc_mvd9x8r","responsibility":"government","reasoning":"virtue","policy":"none","emotion":"outrage"}, {"id":"rdc_mvae20f","responsibility":"company","reasoning":"consequentialist","policy":"industry_self","emotion":"resignation"}, {"id":"rdc_mvat949","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"}, {"id":"rdc_mvcb9rq","responsibility":"user","reasoning":"unclear","policy":"unclear","emotion":"indifference"} ]