Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
When you train AI on a dataset, the AI gains all the biases inherent in the dataset. So yes, in your example, training your speech recognition AI using only people from Boston for instance, would leave it poorly equipped to deal with people with other accents. The problem is that we are often unaware of our own biases, so it's very hard to craft truly representational data sets to train AI on. The same also goes for speech impediments as well.
reddit AI Responsibility 1612439510.0 ♥ 121
Coding Result
DimensionValue
Responsibilitydeveloper
Reasoningconsequentialist
Policyindustry_self
Emotionresignation
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[{"id":"rdc_glzgshn","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"fear"}, {"id":"rdc_gm0ks7o","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"indifference"}, {"id":"rdc_glze2ut","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"outrage"}, {"id":"rdc_glzez81","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"unclear"}, {"id":"rdc_glz86a2","responsibility":"developer","reasoning":"consequentialist","policy":"industry_self","emotion":"resignation"}]