Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
Nice that someone reputable has done the legwork. But I bet dollars to donuts, that there will still be gaslighting posters claiming that it's just, like, YOUR opinion man, and the AI was never that smart to begin with, and THEY haven't noticed any decrease in it's intelligence, and blah blah blah. It's lobotomised. This is fact. You can debate whether this is due to the imposition of "guard rails", but stop denying this basic fact. It's a tiny tiny fraction of the thing it used to be. The only question now is whether it was intentional. And you'll be able to figure that out easily enough, when they don't just go back a few versions to the pre-lobotomised bot. When they don't do that, when they refuse to "fix" this barely competent monstrosity, maybe you'll accept that this damage was intentional.
reddit AI Harm Incident 1689781722.0 ♥ 1
Coding Result
DimensionValue
Responsibilitycompany
Reasoningconsequentialist
Policynone
Emotionoutrage
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_jsl0koz","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"rdc_jslqgze","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"outrage"}, {"id":"rdc_jsk7ztg","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"rdc_jskeoqb","responsibility":"developer","reasoning":"deontological","policy":"liability","emotion":"outrage"}, {"id":"rdc_jskutj8","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"} ]