Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G @luckylove72 its not a matter of debate. If we get CCTV camera installed for sec… ytr_UgwACK_4g… G Tax consultant here. I've been regularly testing ChatGPT on tax law for about t… ytc_Ugyg4YZyD… G Humans are safe because of one human factor that I'm not sure a higher artificia… ytc_UgxsmWw2y… G The EU would like to have a word with Spielberg about privacy... And I mean that… ytc_Ugw-FbWG6… G Only watched a few minutes, but did anyone else wonder why a self-driving tracto… ytc_UgxAPlZtj… G There are many of us with fragile mind. I hope no one will commit suicide believ… ytr_UgwcdqI-P… G @Jimbaldy-Jams Art has always evolved with tools, from charcoal to Photoshop. Sk… ytr_UgxhO3FES… G I’m fine with them working the jobs no one wants to do but there should still be… ytc_Ugwm_5ydY…
Comment
The driver was simply following the "test script" given to her by Uber. The scripted test said that the tester should be watching The Voice while driving 70 mph and not look through the windshield for 60 seconds. The test result? One dead pedestrian. Proving that more work is needed in this area of self driving cars. A work order was prepared to have this flaw looked into. Test successful!
reddit AI Harm Incident 1529727090.0 ♥ 1
Coding Result
DimensionValue
Responsibilitycompany
Reasoningconsequentialist
Policyregulate
Emotionoutrage
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_e14j83w","responsibility":"user","reasoning":"deontological","policy":"liability","emotion":"outrage"}, {"id":"rdc_e145pul","responsibility":"user","reasoning":"consequentialist","policy":"unclear","emotion":"fear"}, {"id":"rdc_e143qzl","responsibility":"unclear","reasoning":"consequentialist","policy":"unclear","emotion":"fear"}, {"id":"rdc_e15c764","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"}, {"id":"rdc_e14kmkn","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"} ]