Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
A thought experiment. Trigger warning for sensitive people One day your car licence expires. You are late renewing the licence because your wife dies in a car accident a week before your licence expires and you’ve forgotten to pay the bill. You go to the government hub to pay the bill. The government has outsourced the customer service at the government hub to an ai. You enter the hub and try to talk with the ai and get your licence renewed but the ai is programmed to invalidate your request for a licence and prevent you from driving again until you pay an unreasonable $10,000 dollar fine. You plead with the ai to be lenient stating your wife has just died in a car accident. But the ai doesn’t listen and keeps repeating you must pay the fine or you will be immediately arrested. You must pick your 2 year old daughter up from child care in 1 hour and you need to drive to get there. You become distraught and you begin acting irrationally and shouting and attacking the desk the ai sits behind. The ai simply repeats the need to pay the fine eventually calling police security bots to arrest you because of your actions. You are brutal arrested, you state the need to pick up your daughter as you are her only carer as you have no family nearby to help. The robot police state you are resisting arrest and you are pinned to the ground eventually loosing consciousness and stop breathing. You die on the way to the police station. Eventually your story makes it to the news and becomes viral. Humans for the most part are irrational when stressed. The logical brain of agi meeting the irrational brain of a human only has one consequence, conflict. The risk of backlash against ai making the wrong call and just following its defined policies will eventually lead to human backlash on agi in general human roles meaning it just won’t be worth deploying in those roles. That’s not to say agi won’t take all the back of house roles but there is likely to always be a human face in front of the machine. It gives comfort and facilitates compliance.
youtube AI Governance 2025-10-19T16:3…
Coding Result
DimensionValue
Responsibilitygovernment
Reasoningconsequentialist
Policyunclear
Emotionmixed
Coded at2026-04-27T06:24:53.388235
Raw LLM Response
[ {"id":"ytr_UgweWb8V6ooN4y26sY94AaABAg.AOUjEH68-gwAOWL9fLupj","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"}, {"id":"ytr_Ugxfky1qm7lohqVZ0-l4AaABAg.AOUNhKLXfMHAOg2yGhTqtv","responsibility":"ai_itself","reasoning":"deontological","policy":"none","emotion":"fear"}, {"id":"ytr_UgwHMluy0kJn5blLn594AaABAg.AOTbghk7cYHAOhWGXvZGCO","responsibility":"ai_itself","reasoning":"deontological","policy":"none","emotion":"fear"}, {"id":"ytr_Ugy_douuXaThiwuE12t4AaABAg.AOT84Y_skOpAOUHh2jY8bc","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"ytr_UgzfAJdivTacc1Bv8hJ4AaABAg.AOSngdo6THGAOSpBzrPLAf","responsibility":"unclear","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"ytr_UgzLDhXF2__Ga3la0-R4AaABAg.AOSVutMQEQHAOTMQFc55r5","responsibility":"government","reasoning":"consequentialist","policy":"unclear","emotion":"mixed"}, {"id":"ytr_UgxK-3vCrjGW3OF-uGh4AaABAg.AOSJAeB8fDTAOSJsmxwKJe","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"fear"}, {"id":"ytr_UgzEeQHIuuEBPWkhNWp4AaABAg.AOSE63l0xAIAOSIZ3qnkF5","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"mixed"}, {"id":"ytr_UgzEeQHIuuEBPWkhNWp4AaABAg.AOSE63l0xAIAOSJGxJPPd3","responsibility":"unclear","reasoning":"mixed","policy":"unclear","emotion":"mixed"}, {"id":"ytr_Ugw0CdOPYqMVe6_xjZ54AaABAg.AORz_gMu4AiAOS2mV0toh","responsibility":"unclear","reasoning":"consequentialist","policy":"unclear","emotion":"indifference"} ]