Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
LLM's should never be used to evaluate résumés. It is as easy as that. There are way too many problems atm: 1. As mentioned the LLM will have bias from its training data 2. Future training data can get polluted by the LLM's reinforcing the bias. 3. It is a black box. We can't explain how it made the decision so how can HR justify using it? What if they are accused of breaking the law? 4. If these LLM's get popular companies start hiring using very similar patterns. If discrimination is bad now imagine all companies discriminating in the exact same way against the same people. We could end up with some people being unable to get a job for having the wrong name or date of birth 5. Companies can and absolutely will get accused of forming a cartel because they use the same LLM. It isn't that different than directly working together for price fixing instead it happens on the hiring side. There are probably many more issues ...
reddit AI Bias 1730492102.0 ♥ 157
Coding Result
DimensionValue
Responsibilitycompany
Reasoningdeontological
Policyban
Emotionoutrage
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_luwvnwc","responsibility":"company","reasoning":"deontological","policy":"ban","emotion":"outrage"}, {"id":"rdc_luwz8l5","responsibility":"user","reasoning":"consequentialist","policy":"industry_self","emotion":"approval"}, {"id":"rdc_luzsin6","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"rdc_lv1gtup","responsibility":"distributed","reasoning":"mixed","policy":"none","emotion":"resignation"}, {"id":"rdc_luxtq7d","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"approval"} ]