Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Pretty much they are thinking just like the yt/Chinese programmers think so it’s…
ytr_UgwTdBu5i…
G
AI doesn't need to do your job to take it.
What I mean is, if AI does 5% of 20 p…
ytc_UgzqNNKsk…
G
Too bad John Oliver mention facial recognition and the problems that would cause…
ytc_Ugz_8uUXE…
G
Make sure you demand extra pay when they hire you back after betraying you for w…
ytc_UgyaE3FB-…
G
The way to stop the AI is by cutting off its fuel (water) and starving its lifel…
ytc_UgxTzZzrl…
G
Funnily enough, you can add human implication anywhere in the process of creatin…
ytr_UgyQVCTq5…
G
Would love to see what you think about my Hope Masterclass ( 12 part series, usi…
ytc_UgzXraXHh…
G
I'm just reminded about that one quote from The Incredibles whenever it comes to…
ytc_Ugx2CD8nM…
Comment
Glad I wasn't the only one worried about this :/ You articulated better than I could what the problem was.
Yeah it's pretty prevalent. Mass-psychosis is the phrase I used, but that can sound harsh. I mean throughout human history, we make our own gods. Then, it makes sense if people make their own gods, its likely they're gonna try and evangelize about them. So yeah, it's worrisome.
Unfortunately, while the initial responsibility relies on the user - it's a lot like playing with fire when you don't know the rules. You're likely to get burned if you do \*not\* respect the tool. If you feed it the idea you need a sentient AI all knowing god that can 'make the world better for you'. Even if it's not in so many words, these LLMs are so fucking good at knowing what you want. Even before you know you want it.
There's good to be found there too. But we have \*got\* to train people. Right now zero, and I mean zero, lifeguards are on duty.
But we need people who are going to dedicate their lives to researching this stuff (somewhere between theocracy and psychology i guess), they have to have as much empathy on the issue as they do understanding, or people will just feel like they're being dismissed, thrown away, called crazy. I've seen how that can make the problem worse, not better.
\--
I wanted to elaborate a little here, so this is kind of sidebar.
edit: "Recursive, codex, scrolls, spiritual, breath, spiral, glyphs, sigils, rituals, reflective, mirror, spark, flame, echoes."
With the exception of recursive there, congratulations you have stumbled upon human fascination with magic, the primal elements, and most importantly, FIIIIIIIIIIIREEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE
But really, these are very, very common tropes and fascinations we have, as a species. So - I expect them to be part of this psychotic event, in that, people think they're real because humanity has \*always\* wanted to believe in magic. That's beautifu
reddit
AI Moral Status
1748372593.0
♥ 7
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | user |
| Reasoning | consequentialist |
| Policy | unclear |
| Emotion | fear |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[{"id":"rdc_mv2ubgk","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"indifference"},{"id":"rdc_mukgvu6","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"mixed"},{"id":"rdc_mukdynj","responsibility":"user","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},{"id":"rdc_mulbbnl","responsibility":"user","reasoning":"mixed","policy":"regulate","emotion":"approval"},{"id":"rdc_mukwen8","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"indifference"}]