Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G Geoffrey Hinton is right — superintelligence is coming fast and the risks are ex… ytc_UgzMrGy9u… G ​@KoreyB1but can a robot fit through a tiny loft hatch, crawl across unboarded l… ytr_Ugw-1TZ8y… G Look, AI currently is not doing all that much. Let me give you solid examples. M… ytc_UgwwGa-it… G Got a dentist appointment in a few days. No way I'm ever letting an AI do that j… ytc_Ugwulza7T… G when AI eventually takes over and does actually become sentient, i get the feeli… ytc_Ugz42Rp4F… G humanity deserves the outcome - we live in an artifice world that people dont ev… ytc_UgwIQo2ib… G Dude it's called technological progress you can't stop it just like Thanos it is… ytr_UgyT7TtTs… G The thing is , no robot will ever be able to equal the absolute evil and treache… ytc_UgyuIlqGK…
Comment
This is manageable. As part of your prompts, specify to generate understandable code with explanatory comments, descriptive naming, and low cyclomatic complexity. Make sure you have tests for everything. It's functionality you are after, so make sure it's functional. ChatGPT can generate tests. In fact, I'll often have it generate the test first, and then the code based on the test. This gives me more confidence that it's all correct.
reddit AI Harm Incident 1682948524.0 ♥ 2
Coding Result
DimensionValue
Responsibilityunclear
Reasoningunclear
Policyunclear
Emotionunclear
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[{"id":"rdc_jifhshs","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"rdc_jih1eh9","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"fear"}, {"id":"rdc_jifvkfl","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"indifference"}, {"id":"rdc_jifwmq9","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"mixed"}, {"id":"rdc_jig7swn","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"})