Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
My guess is, the AI is programmed to look for the move that is most likely to lead to a winning result, and at the moment, it correctly saw that the only way to win is if the opponent makes some mistake, however it does not know which kind of mistake is most likely to be made by the opponent, for that's not part of the AI search algorithm. These move were trying to make the opponent to commit some very simple mistakes, which is very unlikely to happen.
reddit AI Jobs 1457877871.0 ♥ 17
Coding Result
DimensionValue
Responsibilityai_itself
Reasoningconsequentialist
Policynone
Emotionindifference
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[{"id":"rdc_d0y78rb","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"},{"id":"rdc_d0yfp4o","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"},{"id":"rdc_d0yl42f","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"},{"id":"rdc_d0y2pbl","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"amusement"},{"id":"rdc_d0y7us8","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"mixed"}]