Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
We actually don't. So far, tests have been strictly controlled, and in most cases, they have been the only self-driving car on the road. Not to mention that we have mere hundreds of hours of data. We need hundreds of thousands or millions before we can trust them so implicitly.
reddit AI Harm Incident 1475430191.0
Coding Result
DimensionValue
Responsibilitydeveloper
Reasoningconsequentialist
Policyregulate
Emotionfear
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_d8ai0nx","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"}, {"id":"rdc_d8almjm","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"rdc_d8b7vpz","responsibility":"distributed","reasoning":"consequentialist","policy":"unclear","emotion":"fear"}, {"id":"rdc_d8ar0o0","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"mixed"}, {"id":"rdc_d8azx7v","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"fear"} ]