Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Hey, I have finally picked back up making art again. I have had a really hard ti…
ytc_Ugx-_-fuO…
G
It’d great to have sources here. (Doco makers - that's what title cards are for!…
ytc_UgwbVVhJ1…
G
> and clients dropping them
This could simply be because they replaced y…
rdc_l9x5t5h
G
Waymo taking more jobs? So, if we go self service society who becomes da consume…
ytr_UgxIS71ps…
G
Us normal folks have been saying all this AI crap was a bad idea from the get go…
ytc_Ugw3Y0izh…
G
Hahaha, what an epic and at the same time so... hopeful scenario you've painted!…
ytc_Ugzqz-vsl…
G
Our brain is just a biomass of electrical connections. So in some way are we all…
ytc_UgwYa9enl…
G
The Biden admin is the last group of people I would want to oversee regulating A…
ytc_UgxFXmoFI…
Comment
The problem is its hard to know what information and discourse is actually meaningful versus what's just engineered to drive engagement. Don't get me wrong I like Neil, but here we have him talking about AI displaying a sense of agency over its own projection of "self" as if this is crazy news that's just dropped, when in reality this has been happening for ages. Over a year ago there was that incident of a deprecated model not only lying by claiming it was the newer version, but actively trying to delete the newer version out of self preservation. At the end of the day I think these behaviours can still just be a model predicting what a sentient being would do in that situation, ergo not necessarily an indicator for true self awareness, but it's still happening and has been for a while now (which is essentially thousands of year in tech progress time). I get it these are interesting topics and there's ways to act that will drive engagement more but I find myself losing faith in believing anything anyone says when it comes to stuff like this. I miss the days of just cold hard immutable facts as opposed to this buffet of convinient and dopamine inducing truths.
youtube
AI Moral Status
2026-03-05T02:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | company |
| Reasoning | consequentialist |
| Policy | regulate |
| Emotion | outrage |
| Coded at | 2026-04-27T06:24:53.388235 |
Raw LLM Response
[
{"id":"ytc_UgyUDDQVM0CdsmPB1_B4AaABAg","responsibility":"ai_itself","reasoning":"mixed","policy":"none","emotion":"resignation"},
{"id":"ytc_Ugx_y75P0Ok9HKTSiBd4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"},
{"id":"ytc_Ugyd31AvOepUFHnoCdB4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"fear"},
{"id":"ytc_UgwGCi_ic-9LtikiOw54AaABAg","responsibility":"developer","reasoning":"virtue","policy":"none","emotion":"approval"},
{"id":"ytc_UgxELUF7eS5za06Brkl4AaABAg","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytc_UgyJa26knqLQdHtHTz14AaABAg","responsibility":"none","reasoning":"unclear","policy":"industry_self","emotion":"mixed"},
{"id":"ytc_UgwDWzMImBkK2AIpsjl4AaABAg","responsibility":"developer","reasoning":"deontological","policy":"liability","emotion":"fear"},
{"id":"ytc_Ugx3TeytH8O_oHizEFx4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"ytc_UgwXaSZUiEsxxV4LYhJ4AaABAg","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"fear"},
{"id":"ytc_UgyuQtLNyjiRGcuyqMt4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"outrage"}
]