Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
And also the policy on follow up. If this system is automatically launching a drone strike that’s a hell of a lot more concerning than sending a tech an email saying “this looks like a match to me. Is this a match?”
reddit AI Harm Incident 1562196744.0 ♥ 3
Coding Result
DimensionValue
Responsibilitynone
Reasoningconsequentialist
Policyregulate
Emotionfear
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"} ]