Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G Yes, thank you for not being one of those AGI believers. You had me hooked when … ytc_UgzKv5C6b… G Honestly, the butter hands could've been funny, but I (not a coward) would just … ytc_Ugw8-wPJM… G I think that opinion on research uses is interesting but practically, it will re… ytr_Ugxy6CQ_t… G I think the AI bubble.will burst, then there will only be like 3 large.corporati… ytc_UgzbqpI6S… G has AI replaced any job? I have yet to hear about AI having any job consuming im… ytc_UgxQuzsBh… G to cry wolf that some vague, unspecified AI will murder us all in a vague unspec… rdc_kqtxh2q G Have any of you seen I robot the movie an u seen the destruction an disaster tha… ytc_UgzYlJEGQ… G i love how people like to talk about AI playing Chess and Go... i have yet to lo… ytc_Ugw18weYW…
Comment
thats fair and i think for current code review its totally manageable. my concern is more about the trajectory tho. right now you can review every line because the output is code you understand. but anthropic is specifically talking about AI writing code for training future models, not regular software. that research code is gonna get increasingly exotic and harder for humans to meaningfully review even if they read every line. the 90/10 split works today because the 10% you write gives you deep understanding of whats happening. question is whether that holds when the AI is optimizing things humans didnt design in the first place
reddit AI Moral Status 1773273665.0 ♥ 4
Coding Result
DimensionValue
Responsibilityunclear
Reasoningunclear
Policyunclear
Emotionunclear
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[{"id":"rdc_oa2bwfk","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"fear"}, {"id":"rdc_o9zl5cw","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"rdc_o9vtexd","responsibility":"ai_itself","reasoning":"consequentialist","policy":"regulate","emotion":"fear"}, {"id":"rdc_o9wluvn","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"approval"}, {"id":"rdc_o9y8h8g","responsibility":"ai_itself","reasoning":"consequentialist","policy":"regulate","emotion":"fear"})