Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G Bro when i get mad at chatgpt he always fight back, being ruder and more passive… ytc_UgyYdqA1s… G Given language is fundamental to thinking and learning, do LLM that use differen… ytc_UgyMTZNo7… G I learn and build my knowledge of the world, thus forming my opinions, from a wi… ytc_Ugw7H_5wZ… G Its time humans must invent a ‘secret code’, but then AI will discover the patte… ytc_UgzftbTWM… G Actually..you know what...we artists should make different versions of this AI "… ytc_Ugyz2td9H… G The criminals: on no what are going to do😱 there leader: let me handle this „pul… ytc_UgxL8mgYR… G My first thoughts starting reading the sentence was it was shut down because it … rdc_mwvj3pk G In ground heat pumps in the north is frequently not viable. It'd be extremely ex… rdc_gtcsmg8
Comment
Actually I'm glad you asked. I've done programming. I've worked with for and around the guys that you're talking about. And I'm not kidding when I say most of them are white. And computers don't do predictable things necessarily when interpreting data. They can use some pretty unreliable data sets based on those interpretations. For instance when one AI was trying to be trained to look for a husky weird results kept coming up. When programmers looked at the data they found that one of the determining factors the AI was using was snow in the picture. It caused the AI to return results that were Huskies but also other things. Again computers will get weird results trying to please the programmer. If it is being taught to recognize faces and the majority of faces that it is shown are white because most of the programmers are white there becomes a bias and a slant. Hence the racist water fountains and faucets etc. You see? It probably wasn't even intentional originally. But then they started classifying all the codes as top secret and nobody could look at them to determine what the AIs were using to determine these data sets.
youtube AI Harm Incident 2021-04-30T22:2… ♥ 2
Coding Result
DimensionValue
Responsibilitydeveloper
Reasoningvirtue
Policyunclear
Emotionmixed
Coded at2026-04-27T06:26:44.938723
Raw LLM Response
[ {"id":"ytr_UgySbY0fAd9SNaxNmwN4AaABAg.9MtIpwoI8Fp9MxxMi8FBHq","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"}, {"id":"ytr_UgySId0lK89iDs8Ao-h4AaABAg.9MsrYiD0ZZT9MxxbYNcf-Y","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"}, {"id":"ytr_UgzkXQJGEkLd2iwNVu14AaABAg.9Mo5-IC8dJE9MoUXSMccz6","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"}, {"id":"ytr_UgxM6QgoALKPlYQWtY94AaABAg.9MnQlohfjsU9MnyGjQeNBq","responsibility":"government","reasoning":"deontological","policy":"regulate","emotion":"outrage"}, {"id":"ytr_UgwQsMhrSx1PMk_DJVJ4AaABAg.9Mm2qG3INtL9Mm8YxEGetD","responsibility":"developer","reasoning":"virtue","policy":"unclear","emotion":"mixed"}, {"id":"ytr_UgwQsMhrSx1PMk_DJVJ4AaABAg.9Mm2qG3INtL9MnlXY2gvwV","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"}, {"id":"ytr_Ugz_jVhKTKvnS-Ghcfl4AaABAg.9Mm2orVx_IH9Mm6_C8fWKe","responsibility":"ai_itself","reasoning":"consequentialist","policy":"liability","emotion":"resignation"}, {"id":"ytr_Ugz_jVhKTKvnS-Ghcfl4AaABAg.9Mm2orVx_IH9MvkGRvzYB-","responsibility":"government","reasoning":"deontological","policy":"regulate","emotion":"indifference"}, {"id":"ytr_UgxTvmCc1edYDTxFyF54AaABAg.9Mm2T7d9HfO9Mm46_DosUn","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"}, {"id":"ytr_UgxTvmCc1edYDTxFyF54AaABAg.9Mm2T7d9HfO9Mt2IT-z_Zw","responsibility":"developer","reasoning":"virtue","policy":"unclear","emotion":"mixed"} ]