Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Please, please, please, Everyone, especially those making important decisions re…
ytc_UgxP3orvh…
G
Body Language Analysis No. 4249: Will Smith, Sophia the Robot, Robot-Empathy, an…
ytc_Ugz23F41f…
G
Unpopular opinion: generating art with ai is a completely fine and okay thing to…
ytc_Ugw4klZZZ…
G
I think the issues is the fact that system even allows it.
There should be pla…
ytr_Ugwtvm4o8…
G
Humanity is addicted to fantasy ... Terrified of reality ... Weak men = hard tim…
ytc_UgwCULcJ3…
G
Agreed but coming for people for using ai as a selfie and insinuating they put i…
ytc_Ugz0CJp57…
G
AI is only good on things that it's been trained on with high amounts of data. S…
rdc_mle6l7f
G
Hack the automnous robot network, shut it down. See how well the offshore staff…
ytc_Ugwq_x2_f…
Comment
It’s the same thing with testing for a rare illness, with some racial bias thrown in.
In short, if you want to detect something that is rare with any accuracy you need an extremely good test. See the base rate fallacy. Most people don’t view facial recognition in this way and just think a hit is good cause for investigation.
Further, a lot of facial recognition algorithms perform way more poorly on people of color, amplifying this problem. It is quite difficult to remove biases and issues from AI tech, since any racial bias in the training set is learned by the algorithm.
edit: since the wikipedia article has a (poorly written) [terrorist example](https://en.wikipedia.org/wiki/Base_rate_fallacy#Example_3:_Terrorist_identification) you could also view this from the perspective of the fourth amendment: is it acceptable to be subject to a search where the main cause is a test that is less than 1% likely to be right?
reddit
AI Harm Incident
1563715854.0
♥ 3
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | consequentialist |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_euddy2g","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"rdc_eudeyzp","responsibility":"government","reasoning":"deontological","policy":"ban","emotion":"fear"},
{"id":"rdc_eudetdn","responsibility":"government","reasoning":"deontological","policy":"regulate","emotion":"outrage"},
{"id":"rdc_eudf7y0","responsibility":"government","reasoning":"deontological","policy":"ban","emotion":"fear"},
{"id":"rdc_eudfq4v","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"}
]