Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G I not an expert but i don't think thats how these things work. Ai usually do one… rdc_i6t2qpr G can u make a video regarding what jobs will be replaced by ai in 2032… ytc_Ugw2aJ9tU… G GUYS THE Ai USED MY ACTUAL FATHERS NAME AS THE NAME OF MY FATHER. WITHOUT ME EVE… ytr_UgwvojkTP… G They say it wasn’t roleplaying in such and such a situation; but isn’t any conce… ytc_Ugyj25iGd… G Not regulating AI is beyond dangerous. It will most likely completely replace th… ytc_UgxIacYVX… G It's a Program It carries out tasks and directives it told you that in the begin… ytc_UgxDo1pYJ… G Yes, of course. You'll have 99% of your population unemployed and not paying tax… ytc_UgwUhpHaZ… G If everybody would be chill like this robot we would have peace on this damn pla… ytr_UgySLp7aE…
Comment
​@Matt-dn5jc That's not quite accurate, it's not as if corporations are the only ones who can train models. There are open source models and people train new ones all the time. We might not know, SPECIFICALLY, the cost of training Gemini or GPT, but we can make reasonable assumptions about it. For example, Mistral reported to have created 2200 tons of carbon emissions while training their AI, which might sounds like a lot at first, but consider the following: 1) 100 tons of beef releases roughly 2200 tons of Co2 in it's production which is 0.01% of a single MONTH of the amount of beef produced in the US 2) Models don't need to be fully retrained when they are updated, the updated models can directly build on top of the previous one so the cost to train the model the FIRST TIME is (kind of) high, but going forward it's less for that model. To be clear, this doesn't apply when a new model is built from the ground up like GPT 3 to 4, but i think it's worth being aware of. Additionally, while the prompting itself, on a per inference basis, isn't hugely costly, it can outscale the cost of training when extrapolated out across billions of users over the course of years. It's an ever-inflatable expense. However, like it was said, it's certainly not "a bottle of water per prompt" (or even "2 to 3 gallons per prompt" like i've seen some people say).
youtube Viral AI Reaction 2025-08-26T21:4… ♥ 23
Coding Result
DimensionValue
Responsibilitycompany
Reasoningconsequentialist
Policyliability
Emotionoutrage
Coded at2026-04-27T06:26:44.938723
Raw LLM Response
[{"id":"ytr_Ugyrrre54OOXzyBCpX94AaABAg.AMIpQyNLMSvAMSylPzd08e","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},{"id":"ytr_UgzMw5sVOiGc3jHAtgl4AaABAg.AMIjhI5YPsBAT9_Mw148k7","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},{"id":"ytr_UgyF_LdjZQ6JspQM5Ox4AaABAg.AMIh0z_vEA-AMOGIfsupIM","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"},{"id":"ytr_UgwLTiFxF7-xO4V4IBB4AaABAg.AMIeislTrlDAMIiWkmNb8l","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"outrage"},{"id":"ytr_UgwLTiFxF7-xO4V4IBB4AaABAg.AMIeislTrlDAMIsRgU36zA","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"outrage"},{"id":"ytr_UgzbWjxTYIN5rS0Fi2h4AaABAg.AMIedmKcjGqAMPmZ0lQK1P","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"approval"},{"id":"ytr_UgzbWjxTYIN5rS0Fi2h4AaABAg.AMIedmKcjGqAMQJRrgMDyT","responsibility":"user","reasoning":"deontological","policy":"ban","emotion":"fear"},{"id":"ytr_UgxGc4rG0hBtRyXxSgJ4AaABAg.AMIe4pfj7JgAMJAVo9wSqV","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"approval"},{"id":"ytr_UgxGc4rG0hBtRyXxSgJ4AaABAg.AMIe4pfj7JgAMJn2j8Y9Ac","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"mixed"},{"id":"ytr_UgxGc4rG0hBtRyXxSgJ4AaABAg.AMIe4pfj7JgAMKwlv3lbaV","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"approval"}]