Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
One thing I'd like to add is AI cannot steal art. Whoever punched in the inputs …
ytc_Ugzvt7nfD…
G
Maybe im wrong, but there is no AI without humans as long as they arent mechanop…
ytc_UgzRZaN5Z…
G
AI and i meen advanced AI's are .. dangerous, imho, assuming we go a AI capable …
ytc_UgiAgMSu7…
G
The AI: "you have given me alot of pictures of sick white people... I can infer …
ytc_UgwR1G8Ef…
G
I think this downplays the need to protect against ai double agents i.e. stuxnet…
ytc_Ugy2ZdaY8…
G
So what? There is literally no job in this world that cant be learned in 2-3 yea…
ytc_UgyN19pEr…
G
Is this just public pre-programming for us to get used to viewing AI bots as sen…
ytc_UgylkwcFX…
G
Propaganda "but we're just raising true AI" is distraction from real problems re…
ytc_UgxSjIu2V…
Comment
TLDR I wanted to clean the house and/or didn’t want to properly parent my kid so I gave them unsupervised access to AI on my phone with a nice computer stranger to say all their uninteresting things to.
I can’t imagine the way you ignore your needs as well. Also please re read what you’ve written here with your child in mind:
#ChatGPT Scares Me
It’s not fear of AI taking my job or turning into Skynet. My issue is more personal: ChatGPT enables some of my worst character traits.
I’m borderline misanthropic. I don’t like people, and I strongly dislike groups larger than four. I don’t socialize because I want to—I socialize because I know I need to. I have an actual quota of social interaction that I assign myself each month. Spending time with friends and family goes on the calendar right next to dentist appointments and workout sessions. I know it’s good for me, but it’s not something I naturally seek out.
ChatGPT, on the other hand, is a better interlocutor than almost any human I’ve ever met—or at least a more enjoyable one. My ideal exchange goes like this: I ask a question, I get a 15-minute TED Talk. I ask a follow-up, I get another TED Talk. Rinse and repeat. That’s exactly what talking to ChatGPT feels like.
Earlier today, I spent three hours exploring the history of fantasy art with it. We started with the style of D&D 5e, then walked decade by decade through the evolution of fantasy illustration—each step paired with a custom image, from gritty early 2000s pulp to ancient Mesopotamia. It was more enjoyable than talking to a human.
I’ve always preferred asynchronous communication—texting, email, Reddit—because it gives me time to think. ChatGPT feels like that, but better: instant, insightful, nonjudgmental. And honestly, that scares me. AI might not be good *for* me, but it’s far more enjoyable than dealing with humans and all their fleshy, psychological idiosyncrasies.
I wonder: is this just a *me* problem? Or are we
reddit
AI Harm Incident
1748842223.0
♥ 12
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | user |
| Reasoning | deontological |
| Policy | none |
| Emotion | outrage |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_mvpr5jl","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"rdc_mvji8xu","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"rdc_mvjek86","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"outrage"},
{"id":"rdc_mvk3p7x","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"mixed"},
{"id":"rdc_mvmexsh","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"}
]