Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G the experiment was to test how fast the algorithm would learn the bots interest … ytr_Ugwtvm4o8… G The global elites are trying to find easier methods to control people. AI is one… ytr_UgzPNL5P8… G I understand your feelings about AI—it can be a polarizing topic! In the video, … ytr_UgyxEPnsp… G This is why it be way cooler if the future was more like starwars, than star tre… ytc_UgxCuy3I-… G Yeah, people who think that DARPA doesn't already have sweet recognition algorit… rdc_ektnt31 G A can prove Bard is intelligent and sentient- I have copies of all conversations… ytc_Ugzf0QeIP… G And what would the USA banning it accomplish exactly? Development would just go … ytr_Ugx4ap0mm… G And now they have AI to help their companies stay profitable and to fight all co… ytr_Ugz1Lzr9c…
Comment
As an amatuer (but serious) AI safety researcher I can tell you AI is usually extremely accurate for well established facts with lots of discussion in academic writing but the accuracy steadily decreases with more obscure or debated things. Additionally, modern Large Language Models are often highly optimized to mirror the user and be agreeable. This means that if at all possible to agree with the user, they will. These factors makes them extremely unreliable for the use case presented in this video. Going to AI to validate your own (non scientifically supported) opinion on a hyperspecific health choice is extremely dangerous. It likely give you the validation that you wanted, but it won't be because you were right.
youtube AI Harm Incident 2025-12-17T03:4… ♥ 2
Coding Result
DimensionValue
Responsibilitydistributed
Reasoningmixed
Policyregulate
Emotionfear
Coded at2026-04-27T06:24:53.388235
Raw LLM Response
[ {"id":"ytc_UgwderVSkp_hvGdACJJ4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"fear"}, {"id":"ytc_UgzvCSjZAlgS0vOjYb14AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"}, {"id":"ytc_UgzpN-uvz-m5w3TX3dd4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"mixed"}, {"id":"ytc_Ugz8c0pqBuv86B5o4nF4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"industry_self","emotion":"approval"}, {"id":"ytc_UgwhKi8u2AocnM9txmh4AaABAg","responsibility":"user","reasoning":"deontological","policy":"liability","emotion":"outrage"}, {"id":"ytc_UgyukuUWV35wHY9rPSJ4AaABAg","responsibility":"distributed","reasoning":"mixed","policy":"regulate","emotion":"fear"}, {"id":"ytc_Ugxvl5Twvqs4LZa2fiN4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"indifference"}, {"id":"ytc_UgzUqDcYkIVQe_2XnZ54AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"}, {"id":"ytc_UgxuyVH2tnw7jctwghp4AaABAg","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"resignation"}, {"id":"ytc_Ugx7YR6HJqIRbmJk_aN4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"mixed"} ]