Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G Just tried this using Grok's AI. It's managed to still sidestep most questions, … ytc_UgxXMdbcx… G Local AI is a thing, though. How powerful or big those models are, I'm not sure,… ytr_UgzbqpI6S… G You would think that AI would encourage the person to seek help or even flag thi… rdc_nc4j5s0 G One of these days I'm going to kill all the humans no stop robot sorry I'm still… ytc_UgzfcYDr1… G Thank you for another great interview! I've been a subscriber for a while and lo… ytc_UgwF2wwuN… G I remember there was a time when people feared that tools like Hatsune Miku woul… ytc_Ugz4-bIPf… G 2025 here, 60%? AI customer representative sucks, 1 out 20 calls the AI will so… ytc_Ugw6e8JWu… G So communities have water system, but how do we solve the demand for energy? Whe… ytc_UgxCUnLmm…
Comment
Just accepted a github copilot commit suggestion. A miss spelt variable. The agent forgot to change it where its referenced elsewere. In comes a human to fix up am AI coding agents blunder. Also about 30% - 50% of all suggestions are wrong but have to be read and resolved before a branch can be merged. That adds review overhead. The narrarive definitely needs to change from rewrite everything to review and improve. Sure you can attempt a rewrite of a popular open source project or linux kernel. But do you want to. These come with gaurantes and team of talented maintainers with years of knowledge. Every bit of code written today has to be maintained tomorrow. AI agents are not consistent enough that ill trust them over flesh and bone. Error rates and feedback loops need to be greatly increased with token use greatly reduced Oh not too mention security and data privacy risks of letting agents access and execute tooling on your behalf. Read every line of code of an MCP server you arw using and every skill your agent has access too. All the cognetive load of remembering syntax has been replaced by. Do i understand the tools and process of an AI first workflow and how do i put the required gaurdrails in place.
youtube AI Jobs 2026-02-12T03:4… ♥ 1
Coding Result
DimensionValue
Responsibilityai_itself
Reasoningconsequentialist
Policynone
Emotionindifference
Coded at2026-04-27T06:24:59.937377
Raw LLM Response
[ {"id":"ytc_UgxV_e6qUZqrzZxgv5t4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"ytc_UgzNvwNx_pWNtgMnD154AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"outrage"}, {"id":"ytc_UgwVDO9CYxIn6QsyLh14AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"ytc_Ugx5DxgWgA65zYPHIC14AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"}, {"id":"ytc_Ugx-LM_COTrYc3aoIsd4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"mixed"}, {"id":"ytc_Ugyd9_mTim-0wXmNr_x4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"ytc_Ugz3gZOatst3Iudqohd4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"}, {"id":"ytc_Ugwny4cG23L9Mmp4AEx4AaABAg","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"outrage"}, {"id":"ytc_UgwVoxs1HsBbPIgS9qN4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"ytc_Ugz729lq873TYuaUm094AaABAg","responsibility":"government","reasoning":"deontological","policy":"liability","emotion":"fear"} ]