Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G art is a skill, It is learned and practiced. take pewdiepie for example, AI "art… ytc_UgwPZwYlq… G the solution is quite simple. tax the robots. for every robot at Amazon or Tes… ytc_UgxLBKlZi… G God put enmity between humanity and cains descendants of lucifer who fell on Mt … ytc_UgxaV0c-v… G Funny how you never see one of these autonomous vehicles driving in the snow or … ytc_UgwNhGthf… G We can't "grant consciousness to AI" because we don't know what consciousness ev… ytc_UgyxTP0U1… G Worked his adult life making AI ( and ton of cash) and he just figured out that … ytc_UgyDY26fe… G OK so hear me out… we get a conspiracy going… that makes every single professor … ytc_UgyRxuxtt… G It’s not true that AI likes white people more than POC, but when you ask an AI i… ytc_Ugx3j12G-…
Comment
What I take from this is not a fear that AI will's choices will not align with what we want from it but a feeling that AI is just not ever going to be reliable. At what level of responsibility can we let these things operate without constant supervision? Can we let them track PTO for employees? Payroll and benefits? Right now, if I ask one of these LLM's to write my e-mail in a more organized matter, it does a great job of formatting it. It also adds some features and persuasive item descriptions to the e-mail that didn't exist. I certainly use AI but it's a strange, super verbose and eloquent, but also kind of stupid assistant. How many policies and documents has this Trump administration thrown out there that just reek of AI generation? Anyone who has used LLM's very much has stories like this. These things have to become more capable and consistent in order to justify the incredible resources being pumped into them. The alien nature to them that the guest describes is very, very real. But, I'm not sure it's dangerous so much as just not really good at interacting with us. How will they slot into the paralegal, HR Manager, bookkeeping roles they seem so ideal for if we cannot afford to stop babysitting their output all the time?
youtube AI Governance 2025-10-16T16:0…
Coding Result
DimensionValue
Responsibilityunclear
Reasoningdeontological
Policyregulate
Emotionfear
Coded at2026-04-26T23:09:12.988011
Raw LLM Response
[ {"id":"ytc_UgxvBmwYTGkXyeYW6-R4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"approval"}, {"id":"ytc_UgxomUtLAIFli9qDizN4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"indifference"}, {"id":"ytc_Ugwv-PyBIIo55woVRY54AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"mixed"}, {"id":"ytc_UgzlF3fOA3y5jCw1gol4AaABAg","responsibility":"unclear","reasoning":"consequentialist","policy":"unclear","emotion":"resignation"}, {"id":"ytc_Ugw0kA53dz4OwbVxmb94AaABAg","responsibility":"user","reasoning":"virtue","policy":"unclear","emotion":"fear"}, {"id":"ytc_UgzBnEP-vckjLOWN-tV4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"industry_self","emotion":"outrage"}, {"id":"ytc_Ugxrg0ddp1-PpuRQ6Xp4AaABAg","responsibility":"unclear","reasoning":"deontological","policy":"regulate","emotion":"fear"}, {"id":"ytc_UgzvXk7USAoGTCexu3p4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"outrage"}, {"id":"ytc_UgzRN5J-ylIOrDGIKnR4AaABAg","responsibility":"user","reasoning":"virtue","policy":"unclear","emotion":"fear"}, {"id":"ytc_Ugw9Lk-Vusvl4pZjN_J4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"outrage"} ]