Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G We do not need AI having access to our Nuclear weapons. What if it decides we ar… ytc_Ugwk_77q1… G ​@notstevebuscemi Self preservation may not necessarily imply consciousness. The… ytr_Ugwtodw50… G Ai art, does it discourage me? No. I am already my own discouragement. Aside fro… ytc_UgyN1eShu… G Shad's a loser who has stagnated in his art and instead of trying to get better … ytc_Ugx2koC_4… G The problem of super intelligence that no one mentions _ Its cost... itll quite … ytc_UgxFdCodp… G Because eclipsing ChatGPT is what humanity needs right now, or because Google is… rdc_jpm2rwf G As an AI artist here is my AI art process: -Click "generate" -beautiful image of… ytc_Ugw9vaxZn… G As a poet, AI is pretty crazy. I would like to think nothing can replace human e… ytc_Ugxli4QIJ…
Comment
As a developer who tested Claude/Cline extensively over more than a year, I can say with confidence: Once your project grows beyond ~5,000 lines of code and spans multiple files, Claude loses structural coherence. It starts forgetting context, mismanaging dependencies, repeating errors, and hallucinating logic. Debugging becomes circular. Refactoring? Unreliable. And this isn’t just Claude. No LLM today – including GPT-4, Gemini, Devin – can consistently maintain architectural integrity across a large codebase. There's no persistent memory, no full graph awareness, and no real understanding of software design patterns at scale. The claim that “AI will write 100% of code within a year” is not just wrong – it's deeply misleading. It ignores fundamental limitations of current LLMs: - No true multi-file reasoning - Shallow state tracking - No long-term refactoring memory - No awareness of tech debt or codebase health We’re years away from reliable full-scale AI coding – optimistically 3–5 years, realistically longer.
youtube 2025-07-09T11:2… ♥ 17
Coding Result
DimensionValue
Responsibilityai_itself
Reasoningconsequentialist
Policynone
Emotionresignation
Coded at2026-04-26T23:09:12.988011
Raw LLM Response
[ {"id":"ytc_UgyIWOJqo-xnBt04kxx4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"resignation"}, {"id":"ytc_UgzbpQL9HGhdw6u2oF54AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"fear"}, {"id":"ytc_Ugy2zy5Mc3vZAObXsmx4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"ytc_UgzLXXc8FXlL8s9LCwZ4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"ytc_Ugyz2UwE6cKgAS_OM_V4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"outrage"}, {"id":"ytc_Ugy1fG7PcB4510md08h4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"ytc_UgwkiB5fMten5cl5RIR4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"resignation"}, {"id":"ytc_UgylUOY-HSFMW7FdwYN4AaABAg","responsibility":"company","reasoning":"deontological","policy":"none","emotion":"outrage"}, {"id":"ytc_Ugwm6Z1noEdiK-ZvU0R4AaABAg","responsibility":"company","reasoning":"virtue","policy":"none","emotion":"fear"}, {"id":"ytc_UgxJl9mxmmv8RiOECA54AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"} ]