Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Maybe if pEoPlE oF cOloUr didn't keep saying how whites are so privileged and sh…
ytc_Ugw1YJYqW…
G
You have to make a study to understand that the less you use your brain to solve…
ytc_UgwrYU9Ik…
G
AI can't fix your car, put a new roof on your house, repair your tooth, anythi…
ytc_UgwyC4J8O…
G
I think we need to sequester AI in society to one domain - one ‘task’ it can hel…
ytc_UgwxIuTPe…
G
AI should be used to make life easier for people, not take away things that brin…
ytc_UgzA7gDEW…
G
WOW! Thanks for test driving autonomous driving platforms I won both Tesla/' …
ytc_UgznWTNW1…
G
i would beg to differ, you dont need carless drivers or driverless cars, but car…
ytc_UgyBh9vUc…
G
People tend to overestimate progress in the short term, but underestimate it in …
ytc_UgxaPOl4M…
Comment
I'm somewhere in between. I doubt that they're intentionally trying to be racist, but I also don't think it's merely a reflection of the data set. It seems pretty clear that they've trained the systems to be "sensitive", and applied that training differently to groups they think need more protecting. The programmers, intentionally or unintentionally, created something that reflected their own value system.
Rozado's research -- for me, at least -- seemed to make that pretty clear. In my own experiences with ChatGPT/Bing, it seems pretty clear as well.
youtube
AI Bias
2023-10-26T13:4…
♥ 8
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | developer |
| Reasoning | mixed |
| Policy | regulate |
| Emotion | mixed |
| Coded at | 2026-04-27T06:26:44.938723 |
Raw LLM Response
[
{"id":"ytr_UgwoqKap4qvFnWv4OKd4AaABAg.9wYyUGQrXSa9wxQD5YBeJX","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgyjlqUNvMNWAqTFqNZ4AaABAg.9wTa09BqdECAJVTCn7HCMx","responsibility":"developer","reasoning":"consequentialist","policy":"industry_self","emotion":"mixed"},
{"id":"ytr_Ugw8aHCPnH2XIswDwg14AaABAg.9wQI9DRthnN9wSwOqTdM4S","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"outrage"},
{"id":"ytr_UgysRqcbkDDo-FgL3it4AaABAg.9wPmM5PT4_q9wSwaDfN_Yp","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"outrage"},
{"id":"ytr_UgysRqcbkDDo-FgL3it4AaABAg.9wPmM5PT4_q9wXzJv3-ugD","responsibility":"developer","reasoning":"consequentialist","policy":"liability","emotion":"outrage"},
{"id":"ytr_UgysRqcbkDDo-FgL3it4AaABAg.9wPmM5PT4_qAC54cKrGgXd","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytr_UgzKRTE6oEIr59_Nq1d4AaABAg.9wGHyjLp2hs9wHy4jHrdzV","responsibility":"developer","reasoning":"deontological","policy":"ban","emotion":"outrage"},
{"id":"ytr_UgzKRTE6oEIr59_Nq1d4AaABAg.9wGHyjLp2hs9wKoAbeU422","responsibility":"developer","reasoning":"mixed","policy":"regulate","emotion":"mixed"},
{"id":"ytr_UgzKRTE6oEIr59_Nq1d4AaABAg.9wGHyjLp2hs9wKpLVKiPQ1","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgzKRTE6oEIr59_Nq1d4AaABAg.9wGHyjLp2hsAHtMPH_QoLX","responsibility":"developer","reasoning":"deontological","policy":"liability","emotion":"outrage"}
]