Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G Freaking market opportunity. how about we build a community with rules. a templ… rdc_oi3cbfo G i work in AI/machine learning professionally and still fully support human artis… ytc_Ugzd17FH3… G Art has always been a trade-off between effort and results. Put more effort into… ytc_UgzBQSXLk… G I think these AI doomers are quite over reaching with AI just destroying us outr… ytc_Ugy64pKz9… G Yeah, the work amount argument is really solid. It's also not that easy to get a… ytc_UgwyPIzB7… G @ In what way would you rather they speak to the public? I'm not sure what else… ytr_Ugw47Ee_9… G asking for a friend: what would a .20c sheet of dot stickers do to a tesla robot… ytc_Ugw3iO-Gi… G When people get unemployed by these self driving trucks then they will loot thes… ytc_UgzDga567…

Comment

I love how these comments are either extremely in favour or extremely against AI in coding. Hilarious. The reality is it performs great in many programming tasks you ask it. Claude + cursor combo does wonders. But it's definitely not perfect at all. Especially given in the hands of junior devs, I think it's a bit dangerous. Models tend to hallucinate and sometimes loop infinitely over a question. It also changes unneccessary pieces of codes. Sometimes it doesn't clearly understand a prompt, or only partly does the job and you need to point it out what it did wrong, despite the fact there is a clear input and output example. It remains token prediction, so there will always be a piece of randomness unfortunately. But in all fairness, it does great it in a large majority of cases so far with minimal reprompting. The quality of the prompting and properly contextualizing what you want plays a huge role as well.

youtube 2025-08-07T23:1…

Coding Result

Dimension	Value
Responsibility	distributed
Reasoning	mixed
Policy	none
Emotion	indifference
Coded at	2026-04-26T23:09:12.988011

Raw LLM Response

[
{"id":"ytc_UgwOaYclmw6hzZ9aic54AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"ytc_UgwImB_UQIWaT7aUieN4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"},
{"id":"ytc_Ugz5XKlvRBxYvjBbOKl4AaABAg","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"approval"},
{"id":"ytc_Ugyf83Zqk0RZXi6jazR4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_UgzbDaqE3pquBYHMa7B4AaABAg","responsibility":"user","reasoning":"virtue","policy":"industry_self","emotion":"approval"},
{"id":"ytc_UgxP7z-Nvn2yWQJ0lGN4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_UgxOkCaNsg8y2fhyyx94AaABAg","responsibility":"developer","reasoning":"deontological","policy":"liability","emotion":"indifference"},
{"id":"ytc_UgwIuOVCDESs38qQz2R4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgxQmh8pi6kudxygF-Z4AaABAg","responsibility":"distributed","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugxdw08m8QRVwcMB8dB4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"approval"}
]

Raw LLM Responses