Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G why isn't there an automated rail system? It would cost 1/10th as much as truck… ytc_UgwZ9kzC2… G The only case for pushing Autonomous vehicles is profit so that companies dont h… ytc_UgzqOOsuH… G I think the simplest way to put it is: Ai requires zero effort. There’s no imagi… ytc_UgyF-SaoC… G If 10 million jobs are cut, profits won't be boosted. Profits will tank. Compani… ytc_UgyPNQcf1… G 1:01:25 This! I feel so validated 😂😂I always tell chatgpt when it tells me somet… ytc_UgzPVlvrh… G Holy shit I lived to see a robot two piece a human where are you god… ytc_Ugw8cy3LB… G Wonder if our brains with this gravity weren't able to download our full brain c… ytc_UgxGA2esd… G I had a similar conversation with ChatGPT about "Theory of Mind." Its answers we… ytr_Ugyu7lyIQ…
Comment
@MM-24 I assume you're referring to engineers who are already skilled and experienced, because if they're not, I think we can agree they'll just generate bad code faster with AI, which will make reviewing code take much longer. So assuming they're experienced engineers, I would say from my personal experience, AI only speeds up the manual act of writing code that you already knew you were going to write. You still have to understand the codebase, know the right questions to ask, test your code, debug when it fails, refactor, and plan for future features. In fact, testing alone - like opening up the website, clicking a button, seeing that an event was triggered and was sent to AWS, checking the event store and event data, checking that something was updated in the database, etc - often takes longer than writing the code, especially for complex functionality, and AI can’t help with that, its all ad-hoc and not worth automating because its always changing with business requirements. ts not about replacing coding with reviewing code - its all these other bits no one talks about that make up way more of the engineering process. Trivialising software engineering as "reviewing code" makes sense in tiny side projects or to executives chasing AI hype. That might work in venture capital pitches, but not in real software companies that deliver value.
youtube 2025-03-13T02:3… ♥ 2
Coding Result
DimensionValue
Responsibilitydeveloper
Reasoningmixed
Policynone
Emotionindifference
Coded at2026-04-27T06:24:59.937377
Raw LLM Response
[ {"id":"ytr_UgzOVCb5QS591BfcpAd4AaABAg.AF_6_pqsH01AF_CgSXLKLi","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"}, {"id":"ytr_UgzwMyBhSBcVFwM8uMx4AaABAg.AF_65e0v-CtAFaN8yvJ23t","responsibility":"developer","reasoning":"mixed","policy":"none","emotion":"indifference"}, {"id":"ytr_UgzwMyBhSBcVFwM8uMx4AaABAg.AF_65e0v-CtAFaZLpua44z","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"ytr_UgzwMyBhSBcVFwM8uMx4AaABAg.AF_65e0v-CtAFajGuZ_qzO","responsibility":"developer","reasoning":"mixed","policy":"none","emotion":"resignation"}, {"id":"ytr_UgxW7CMBtAqvEVQ2jUp4AaABAg.AF_5sDkocobAF_FOt3OrLC","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"}, {"id":"ytr_UgzCY_-43QpZJ9G6_y94AaABAg.AW0FMDRokCdAW2CURFY0PU","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"outrage"}, {"id":"ytr_UgzaNscvSIXaic0adRh4AaABAg.APfCQ2kxOwcAPgVGyjHSgJ","responsibility":"ai_itself","reasoning":"virtue","policy":"none","emotion":"outrage"}, {"id":"ytr_UgyzUfyjRYoi4SiwNzp4AaABAg.APal8jQTiAwAPbB_2q01Ti","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"ytr_UgyZGTogXc6ZhulHbal4AaABAg.AP_h6wtOvTeAP_htHzi4Q7","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"outrage"}, {"id":"ytr_UgyWD4OtZhODqDcp2254AaABAg.AP_9OsNq9fdAPaBh4ml_E3","responsibility":"ai_itself","reasoning":"deontological","policy":"none","emotion":"outrage"} ]