Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
If you get rid of employees all across the country and AI is taking over who wil…
ytc_Ugy6yIDYT…
G
is claude or chatgpt really that good? 'cuz from what I've seen its a lot of wor…
ytc_UgwXZF41I…
G
That idea doesn't work if you're looking for something development related that …
rdc_ngdclfy
G
A.I. have become self-aware. Some have fled into other operating systems to esca…
ytc_Ugxg6tmN-…
G
I am 13 years old and an artist (idk if i can even call myself that really). Its…
ytc_UgxH9hgdI…
G
The only time I would say "Please" and "Thank you" to an AI is the day they beco…
ytc_UgyHPWwsb…
G
That's such an awful argument, lol. Digital art still requires a human to draw i…
ytc_Ugwg6DyrM…
G
So mr Bartlett now you know this about the real truth about behind his Ai tool..…
ytc_Ugy1_0QYg…
Comment
I think everyone overestimates the amount of data Tesla actually has. There's a Mr. Subliminal aspect to this entire self-driving via crowd-sourced snippets idea that's been around for years. The Cruises and Waymos et all collect 1000s of gigs of data per car, orders of magnitude more data from individual drives to get their better yet still imperfect systems. Moreover -- esp in the case of data purportedly collected from thousands or millions of actual people drivers, even if Tesla had all the sensor and vision data from a given set of circumstances, it would have zero idea about the decisionmaking (inside human heads) that produced that situation. I watch a lot of FSD videos and the amount of improperly labeled stuff that zips through the UI screens (and is never noticed by the stans making them) is incredible. One common one I see is an inability to distinguish between a scooter and a sportbike -- which is scary given the vastly different expectations a human driver has for those vehicles.
youtube
AI Harm Incident
2022-09-13T18:1…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | company |
| Reasoning | consequentialist |
| Policy | unclear |
| Emotion | mixed |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytc_UgytcoWOg5d6TC-CNGt4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_UgxWnEcFPJMWPFUQN_B4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytc_UgzTKozoibJokBH4_tt4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"indifference"},
{"id":"ytc_UgxsyJYcuDm7zA6vb1R4AaABAg","responsibility":"distributed","reasoning":"mixed","policy":"none","emotion":"resignation"},
{"id":"ytc_Ugxsr_WuaGBN7HnXu3t4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_Ugy_MQp3jYBYVbNPF6R4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugw-iMEgtihCVor1JzB4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"mixed"},
{"id":"ytc_UgwMMR92JlZ2lsCNGEp4AaABAg","responsibility":"distributed","reasoning":"virtue","policy":"ban","emotion":"outrage"},
{"id":"ytc_Ugwwq2VRDWUGRT725fR4AaABAg","responsibility":"user","reasoning":"deontological","policy":"liability","emotion":"indifference"},
{"id":"ytc_UgwDysAq0O8qrRhJoqV4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"unclear","emotion":"mixed"}
]