Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
On one hand, maybe ChatGPT could have some additional safeguards. On the other, how do you make it literally impossible for someone to twist the LLM's arm into saying what you want it to say without making it nearly non-functional? If this guy was met with 2 dozen "Seek help" type responses before he finally got around it. Would that be sufficient to absolve OpenAI of responsibility?
reddit AI Governance 1762490296.0 ♥ 195
Coding Result
DimensionValue
Responsibilitydistributed
Reasoningmixed
Policyliability
Emotionmixed
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_nnkqiao","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"rdc_nnkh5ky","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"}, {"id":"rdc_nnjmfob","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"outrage"}, {"id":"rdc_nnjr1w8","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"}, {"id":"rdc_nnjnmhc","responsibility":"distributed","reasoning":"mixed","policy":"liability","emotion":"mixed"} ]